O EFEITO DE EVENTOS TÉCNICOS ESPECIAIS NO JOGO SOBRE O SUCESSO DOS TIMES DE FUTEBOL PROFISSIONAL: SUPER LIGA TURCA


EL EFECTO DE LOS EVENTOS TÉCNICOS ESPECIALES DEL JUEGO EN EL ÉXITO DE LOS EQUIPOS DE FÚTBOL PROFESIONALES: SUPER LIGA TURCA


THE EFFECT OF SPECIAL TECHNICAL EVENTS IN THE GAME ON THE SUCCESS OF PROFESSIONAL SOCCER TEAMS: TURKISH SUPER LEAGUE


Olcay MULAZIMOGLU1


RESUMO: O objetivo desta pesquisa é examinar o efeito dos critérios de posse, passe e chute no sucesso dos times nos jogos da Super Liga de Futebol da Turquia na temporada 2019- 2020. Foram examinados 612 dados de cada equipe de 306 partidas disputadas por 18 equipes ao longo da temporada. Em todas as partidas, as frequências do número total de chutes dos times, número de chutes no alvo, porcentagem de posse de bola, porcentagem de passes bem- sucedidos, número de passes, número de passes bem-sucedidos, passes importantes e critérios de pontuação foram coletados na página Whoscored. A análise da Anova unilateral de acordo com os resultados da partida (perder-empatar-vencer) mostrou que a média dos chutes no alvo (p <0,001), número total de chutes (p = 0,03), passes importantes (p = 0,03) e pontuação de classificação (p <0,001) foram significativamente maiores na vitória. Nas análises de acordo com a classificação da liga (classificação 1-6, classificação 7-12, classificação 13-18).


PALAVRAS-CHAVE: Análise de jogo. Chutes no alvo. Posse de bola.


RESUMEN: El propósito de esta investigación es examinar el efecto de los criterios de posesión, pase y tiro en el éxito de los equipos en los partidos de la Superliga de fútbol de Turquía en la temporada 2019-2020. Se examinaron 612 datos de cada equipo de 306 partidos jugados por 18 equipos a lo largo de la temporada. En todos los partidos, las frecuencias del número total de tiros de los equipos, el número de tiros a puerta, el porcentaje de posesión, el porcentaje de pases exitosos, el número de pases, el número de pases exitosos, los pases clave y los criterios de puntuación de calificación se recopilaron de la página web de Whoscored. El análisis de Anova unidireccional de acuerdo con los resultados del partido (perder-empatar-ganar) mostró que las medias de los tiros a puerta (p

<.001), el número total de tiros (p = .03), pases clave (p = .03) y el puntaje de calificación (p

<.001) fueron significativamente más altos en win. En los análisis según la clasificación de la liga (rango 1-6, rango 7-12, rango 13-18).


PALABRAS CLAVE: Análisis del partido. Tiros a puerta. Posesión del balón.


ABSTRACT: The purpose of this research is to examine the effect of possession, passing and shooting criteria on the success of the teams in The Turkish Soccer Super League matches in


1 Universidade Mugla Sitki Kocman (MSKU), Muğla – Turquia. Professor Associado da Faculdade de Ciências do Esporte. ORCID: https://orcid.org/0000-0001-5599-280X. Email: olcaymulazimoglu@mu.edu.tr




the 2019-2020 season. 612 data of each team of 306 matches played by 18 teams throughout the season were examined. In all matches, the frequencies of the teams' total number of shots, number of shots on target, percentage of possession, percentage of successful passes, number of passing, number of successful passes, key passes and rating scores criteria were collected from Whoscored’s webpage. One-way Anova analysis according to the match results (lose- draw-win) showed that means of the shots on target (p<.001), total number of shots (p=.03), key passes (p=.03) and rating score (p<.001) were significantly higher in win. In the analyzes according to the league ranking (rank 1-6, rank 7-12, rank 13-18).


KEYWORDS: Match analysis. Shots on target. Ball possession.


Introdução


Um dos problemas mais importantes enfrentado pelos treinadores e analistas dos times de futebol é obter análises e estatísticas em vídeo de seus próprios times e times adversários. A análise notarial dos jogos de futebol é útil para otimizar o feedback para treinadores e jogadores visando a melhoria do desempenho (CARLING; WILLIAMS; REILLY, 2005).

No campo da análise de desempenho do futebol, vídeos de jogos previamente gravados foram amplamente utilizados para monitorar, avaliar e analisar o desempenho das equipes (JAMES, 2006; LIU; HOPKINS; GOMEZ, 2016), mas recentemente os modernos sistemas de análise de vídeo como AMISCO, OPTA e ProZone fornecem um banco de dados abrangente de jogos de futebol (LIU et al., 2013; LIU; HOPKINS; GOMEZ, 2016; MACKENZIE; CUSHION, 2013).

As análises de desempenho dos jogadores associadas aos resultados das partidas podem ser muito úteis para explicar os efeitos no desempenho da equipe (LAGO, 2009; SARMENTO et al., 2014). Muitos estudos baseados em análises de partidas notariais têm se concentrado no número de resultados das partidas (vitória, empate, derrota) e atividades técnicas realizadas pelos jogadores (CASTELLANO; CASAMICHANA; LAGO, 2012; CLEMENTE, 2012; KITE; NEVILL, 2017; LAGO-PEÑAS et al., 2010; LIU et al., 2015).

Foram realizados estudos para associar eventos técnicos e táticos em partidas disputadas em vários campeonatos a nível mundial e continental (Copa do Mundo, Campeonato Europeu etc.) e ligas de futebol com times de sucesso. Além disso, estudos examinando os gols marcados em campeonatos de acordo com vários critérios técnicos e táticos avaliaram o sucesso com base em gols (DURLIK; BIENIEK, 2014; NJORAI, 2014; TOKUL; MÜLAZIMOĞLU,

2018).




Estudos examinando a eficiência técnica dos jogadores de acordo com a linha do placar (vitória, empate, derrota), mudanças durante a partida e o resultado da partida revelaram o efeito do sucesso no final da temporada (ALMEIDA; FERREIRA; VOLOSSOVITCH, 2014; LAGOPEÑAS; DELLAL, 2010; REDWOOD-BROWN, 2008).

Algumas pesquisas avaliaram a frequência de desempenho de atividades técnicas selecionadas em relação ao resultado e ao status da partida nas partidas dos campeonatos. As mais importantes são as frequências de diferentes tipos de passes e maior porcentagem de posse de bola quando a equipe está jogando para mudar um placar desfavorável (derrota e empate) (KONEFAL et al., 2018).

Em um estudo realizado no campeonato espanhol, foi relatado que os times da casa e os times perdedores que sofreram gols tinham mais posse de bola. A identidade do adversário, quanto pior o adversário, maior será a posse de bola (BLOOMFIELD; POLMAN; O'DONOGHUE, 2005; LAGO; MARTÍN, 2007).

Já em um estudo da Copa do Mundo de 1990 e 1994 por Hughes e Franks (2005), foi identificado que times bem-sucedidos produziram mais gols por posse de bola usando sequências de passes mais longas. No entanto, os times malsucedidos adotaram um estilo de jogo mais direto, utilizando sequências de passes mais curtas, já que estas equipes não tinham um nível de habilidade suficiente para manter a posse da bola. (HUGHES; FRANKS, 2005).

Muitos outros estudos de pesquisa também apoiaram a ideia de que os estilos de jogo diferem entre times bem-sucedidos e malsucedidos quando uma variedade de diferentes indicadores de desempenho foi analisada (LAGO-PEÑAS et al., 2011; LAGO-PEÑAS et al., 2010).

Além dessas pesquisas, os espectadores de futebol, torcedores e programadores de TV conseguem obter resultados através de análises de jogos. Algumas estatísticas são refletidas na tela enquanto os jogos são disputados. Informações básicas como a porcentagem de posse da bola, passe total/passe no alvo são exibidos na tela. Comentaristas de futebol avaliam a superioridade dos times em seus comentários enquanto assistem aos resumos das partidas em programas de TV, mencionando tais estatísticas. Nesse sentido, há uma questão importante a ser questionada: Mais posse de bola e mais passes é um sinal de superioridade no jogo? Disparar mais do que seus adversários trará o campeonato ao seu time no final da liga?

O objetivo desta pesquisa é examinar o efeito dos critérios de posse de bola, passes e chute no sucesso dos times em jogos vencedores e no ranking da Super Liga da Associação Turca de Futebol (TFA) nos jogos da temporada 2019-2020.




Métodos


Amostra, recurso de dados e variáveis


A Super Liga da Associação Turca de Futebol (TFA) é o mais alto nível de futebol profissional na Turquia, está entre as principais ligas da Europa. Cada temporada começa em agosto (final do verão na Turquia) e em maio (final da primavera) termina. Em cada temporada, cada time joga contra cada um dos outros times duas vezes, uma em casa e a outra fora. No sistema da liga da temporada 2019-2020, com 18 times na Liga, cada time jogou 34 partidas para um total de 306 partidas na temporada. O ranking de final de temporada é determinado pelos pontos (vitória por 3 pontos, empate por 1, derrota por 0) somados dos 30 jogos de cada time. Todos os 306 jogos da Super Liga TFA da temporada 2019-2020 foram escolhidos como amostra do estudo atual. 34 partidas jogadas por cada um dos 18 times da liga foram avaliadas como participantes separados, e variáveis pertencentes a um total de 612 participantes formaram o conjunto de dados. Os parâmetros obtidos em cada jogo foram comparados de acordo com seu sucesso.

No final da temporada na Super Liga, os cinco melhores times da Turquia tiveram o direito de participar da Liga UEFA. Entretanto, um dos cinco primeiros times perdeu seu direito de participar devido à penalidade da UEFA. Ele foi substituído pelo time da sexta posição. Por outro lado, embora as ligas tenham sido suspensas nesta temporada devido à epidemia do Covid-19, todos os jogos foram disputados. A fim de evitar o impacto negativo da pandemia nos clubes, as três equipes que deveriam ser rebaixadas pela decisão da Federação não foram rebaixadas.

Os dados relacionados ao desempenho desses jogos foram coletados de sites de acesso público “https://tr.whoscored.com” cujos recursos de dados são “Whoscored” apoiados pela OPTA Sportsdata Company (WHOSCORED, 2020).

A confiabilidade entre operadores do sistema de rastreamento da empresa (OPTA Client System) utilizada para coletar estatísticas de jogos de futebol foi identificada por estar em um nível aceitável. Além disso, a confiabilidade do sistema de rastreamento (OPTA Client System) usado pela OPTA Sportsdata Company para coletar estatísticas de partidas de futebol foi testada por Liu e colegas (LIU et al., 2013), o que mostrou que os times que jogam eventos codificados por operadores independentes usando este sistema chegaram a um acordo muito bom (os valores ponderados de kappa foram .92 e .94) (LIU et al., 2013). O estudo foi realizado em conformidade com a Declaração de Helsinque. A aprovação do estudo atual pelo




comitê de ética foi obtida na Universidade Muğla Sitki Koçman (Aplicação nº: 200346). Permissão para utilizar as estatísticas publicadas no site “whoscored” foi obtida.

Em todas as partidas foram coletadas as frequências do número total de remates dos times, número de remates no alvo, porcentagem de posse de bola, porcentagem de passes bem-sucedidos, número de passes, número de passes bem-sucedidos, passes-chave e critérios de classificação OPTA. O sucesso dos times foi analisado de acordo com os resultados da partida (vitória/derrota) e sua classificação (equipes de nível superior: 1-6 classificadas no grupo ‘UEFA’, equipes de nível médio: 7-12 classificadas no grupo inferior: 13-18 classificadas na linha de rebaixamento e próximos times) no final da temporada e o status de “home-away” d time.

Os times são divididos em três grupos enquanto são avaliados de acordo com sua ordem de sucesso. O primeiro grupo foi formado por times que terminaram a liga nos seis primeiros colocados. O segundo grupo foi formado por seis times seguintes, ou seja, times de 7-12. O terceiro grupo era composto de seis times no final do ranking.


Definição dos parâmetros


Chute: Chutes com qualquer parte legal do corpo com o objetivo de marcar um gol, seja ele preciso ou não.

Chute no alvo: Todas as tentativas de gol que podem resultar em um gol se não forem bloqueadas ou salvas.

Passagem: Bolas passadas voluntariamente por um jogador a um colega do time.

Sucesso no passe: os passes voluntários de um jogador são tomados por seu colega de

time.

Posse de bola: A quantidade de tempo que um time controlou a bola durante um jogo,

a partir do momento em que é tomado posse da bola do time adversário sem nenhuma interrupção clara, como uma proporção do tempo total quando a bola estava em jogo (KONEFAL et al., 2018).


Análise estatística


Todas as variáveis foram verificadas quanto à distribuição normal (um valor de curtose foi ±1,0) e homogeneidade de variância (teste de Levene). Foram calculados as médias aritméticas e os erros padronizados. Para comparar os valores médios das variáveis




examinadas, foi utilizada uma ANOVA unidirecional. Assim que encontrado o tamanho do efeito significativo, um teste de LSD pós-hoc de Fisher foi realizado. O nível de significância estatística foi fixado em p < .05.

Além disso, a eta parcial ao quadrado (η2) foi calculada, e os tamanhos dos efeitos foram determinados: tamanho do efeito pequeno (≥0.01 e <0.06), tamanho do efeito médio (≥0.06 e <0.14), tamanho do efeito grande (≥0.14) (Cohen, 1988). Todas as análises estatísticas foram feitas usando o pacote de software SPSS (versão 22.0, SPSS Inc., Chicago, IL).


Resultados


Os dados revelaram que havia uma diferença significativa na média do total de chutes (p =,03) e chute no alvo (p <,001) de acordo com o status de vitória, empate e derrota dos times. Também foi encontrada uma diferença significativa na média de passes-chave (p =,03) e na classificação OPTA (p <,001) por sucesso na partida. Ademais, o teste de comparação múltipla (LSD) revelou: total de chutes, chutes no alvo e passes-chave para o vencedor na partida do que times perdedores e empatados. O “eta parcial ao quadrado” encontrado no chute ao alvo (Tabela 1).


Resultado do jogo

Variável

1 Perder

(n=224)

Empate

(n=164)

Vitória

(n=224)

F(Sig.)

P.eta ao quadrado

(ɳ2)

ós Hoc (LSD)

Tabela 1 – Análise comparativa das variáveis (ANOVA) e tamanho do efeito de acordo com o resultado do jogo (média±sd)



Classificação OPTA

6.36±.20

6.67±.12

7.02±.24

611.41(. 001)**

.668-L

3>1,2, 2>1

Total de chutes

12.46±5.15

12.57±5.36

13.67±4.90

3.73(. 025)*

.012-S

3>1,2

Chutes no alvo

3.65±2.12

4.09±2.16

5.55±2.44

43.20(. 001)**

.124-L

3>1,2

Posse da bola %

50.45±10.55

50.02±10.38

49.55±10.55

0.42(.659)

.001-S

nulo

Sucesso no passe %

79.75±5.83

78.30±6.83

79.69±6.78

2.91(.055)

.009-S

1>2, 3>2

Total de passes

426.75±96.98

416.24±96.25

422.72±103.98

0.53(.589)

.002-S

nulo

Sucesso no passe %

344.06±97.56

330.78±98.22

342.79±106.56

0.95(.388)

.003-S

nulo

Passe-chave

9.46±4.17

9.37±4.46

10.35±4.14

3.46(. 032)*

.011-S

3>1,2

Nível de significância: * p <.05; ** p <.01

ES: tamanho do efeito (ɳ2); S: pequeno (≥.01 e <.06), M: médio (≥.06 e <.14), L: grande (≥.14) Fonte: Elaborado pelo autor





Os dados revelaram uma diferença significativa em todas as variáveis dos times agrupados de acordo com a classificação do campeonato. Os primeiros times classificados em um terço, registraram mais chutes totais (respectivamente: 14,48±5,10; 12,82±5,16 11,50±4,74) e chutes no alvo (respectivamente: 5,18±2,48; 4,34±2,35; 3,88±2,20) do que o segundo e terceiro times em um terço (p<.001). Em termos de porcentagem de posse de bola, os primeiros times de um terço tinham uma porcentagem maior (respectivamente: 53,29±10,15%; 49,22±10,34%; 47,51±10,20%: p<.001). Uma diferença significativa foi encontrada somente entre o primeiro e o último grupo na variável porcentagem de sucesso de passe (respectivamente: 80,50±6,75%; 78,22±6,28%; p=.002). No número total de passes, sucesso de passe e passe-chave, o primeiro terço foi encontrado significativamente maior do que os outros dois subgrupos (p<.001). Enquanto o tamanho do efeito era médio na classificação OPTA e nas variáveis de passe-chave entre os times de sucesso no final da liga, todas as outras variáveis eram pequenas (Tabela 2).


Tabela 2 – Análise comparativa das variáveis (ANOVA) e tamanho do efeito de acordo com a classificação da liga (média±sd)




Ranking da Liga



P.eta ao


Variável


Rank 1-6


Rank 7-12


) Rank 1318

F(Sig.)

quadrado

(ɳ2)

ós Hoc (LSD)


(n=204)

(n=204)

(n=204)




Classificação 6.82±.32 6.64±.35 6.58±.31 28.80(. 001)* .086-M 1>2,3

OPTA







Total de chutes

14.48±5.10

12.82±5.16

11.50±4.74

18.23(. 001)*

.056-S

1>2,3, 2>3

Chutes no alvo

5.18±2.48

4.34±2.35

3.88±2.20

16.09(. 001)*

.050-S

1>2,3, 2>3

Posse da bola %

53.29±10.15

49.22±10.34

47.51±10.20

17.19(. 001)*

.053-S

1>2,3

Sucesso no passe %

80.50±6.75

79.30±6.24

78.22±6.28

6.39(. 002)*

.021-S

1>3

Total de passes

454.01±101.26

408.64±95.14

404.73±94.26

16.30(. 001)*

.051-S

1>2,3

Sucesso no passe %

371.93±105.21

328.29±96.09

319.90±94.34

16.36(. 001)*

.051-S

1>2,3

Passe-chave

11.26±4.20

9.46±4.26

8.56±3.85

22.94(. 001)*

.070-M

1>2,3

Nível de significância: * p <.01

ES: tamanho do efeito (ɳ2); S: pequeno (≥.01 e <.06), M: médio (≥.06 e <.14), L: grande (≥.14)

Fonte: Elaborado pelo autor


Os resultados de testes t independentes revelaram uma diferença significativa entre as partidas dos times em casa e fora, em todas as variáveis do jogo. O desempenho em casa dos times em todas as variáveis foi maior do que fora. Os times jogaram com mais posse de bola (respectivamente: 51,4±10,41%; 48,61±10,41; p<.001) em partidas em casa do que fora, e




jogaram com uma porcentagem maior de sucesso de passe (respectivamente: 80,09±6,28; 78,60±6,61; p=.004). Um número médio de chutes significativamente alto foi encontrado em partidas em casa, e o número médio de chutes no alvo também foi mais alto do que fora. A média de passes-chave foi maior nos jogos em casa. A pontuação da OPTA também foi mais alta em jogos em casa (Tabela 3).


Tabela 3 – Análise comparativa de variáveis (teste t de amostras independentes) e tamanho do efeito de acordo com a equipe da casa e fora



Variável

Início (n=306)

Longe (n=306)


t (Sig.)


ɳ2

Classificação OPTA

6.75±.34

6.62±.34

4.81(. 001)**

.037-S

Total de chutes

14.13±5.14

11.73±4.87

5.94(. 001)**

.055-S

Chutes no alvo

4.9±2.46

4.03±2.26

4.55(. 001)**

.033-S

Posse da bola %

51.4±10.41

48.61±10.41

3.32(. 001)**

.018-S

Sucesso no passe %

80.09±6.28

78.60±6.61

2.86(. 004)**

.013-S

Total de passes

432.9±100.00

412.10±97.72

2.60(. 01)*

.011-S

Sucesso no passe %

351.80±101.17

328.30±99.85

2.89(. 004)**

.013-S

Passe-chave

10.62±4.24

8,90±4,09

5.12(. 001)**

.041-S

Nível de significância: * p <.05; ** p <.01

ES: tamanho do efeito (ɳ2); S: pequeno (≥.01 e <.06), M: médio (≥.06 e <.14), L: grande (≥.14)

Fonte: Elaborado pelo autor


Discussão


Muitas vezes falamos sobre futebol ou criticamos o jogo com nossos amigos depois de uma partida. Antes de mais nada, avaliamos a superioridade dos times em termos dos gols marcados. As posições dos gols dos times, chutes e ‘set play’ são eventos importantes do jogo que são criticados. A evidência científica mais importante que você usa contra seu adversário nesta discussão crítica de jogo serão as estatísticas de jogo, como a taxa de posse de bola dos times, o número total de passes, a porcentagem correta de passes e o número de chutes no gol. Será que as estatísticas do jogo realmente nos dizem que, em uma partida de futebol, uma equipe é superior à outra? As estatísticas de uma partida ou de todas as partidas de uma temporada podem revelar claramente o sucesso do time? Este estudo produziu resultados que responderam a perguntas que as pessoas costumam usar em busca da veracidade dessas informações.




De acordo com os resultados deste estudo, na temporada 2019-2020 da Super Liga Turca, o evento de jogo distintivo mais importante no sucesso das equipas que venceram o encontro foi o número de remates e as variáveis de precisão do remate. Um número significativamente maior de passe-chave foi a variável que explica mais a médias de chutes. Oito eventos de jogos examinados neste estudo revelaram que o sucesso sazonal sustentado é mais importante do que o sucesso de um jogo. Ao contrário do sucesso sazonal (ranking da liga), foi um resultado importante deste estudo que não houve diferença entre as equipes vencedoras, perdedoras e empatadas no jogo tanto na posse de bola quanto na porcentagem de variáveis de passes bem-sucedidos.

Konefal et al. (2018) examinaram a frequência de desempenho de atividades técnicas selecionadas pelos times de futebol que jogaram na EURO 2016 (única fase de grupos) em relação ao resultado e ao status da partida. Este estudo revelou que a atividade técnica mais importante para os times no final de uma partida é o número de chutes no gol e a taxa de sucesso do chute. Entretanto, quando a equipe está jogando para mudar um placar desfavorável (perdendo e empatando em status de partida), o mai importante é a frequência de diferentes tipos de passes e maior porcentagem de posse de bola (KONEFAL et al., 2018).

Lago-Peñas et al. (2018) analisaram várias atividades técnicas em jogos da liga de futebol profissional masculino. As equipes vencedoras tinham médias de chutes (W: 14,4±5,1, D: 13,6±5,2, L: 11,9±4,8) e chutes no gol (W: 6,6±2,8, D: 5,1±2,7, L: 4,2±2,4) que eram

significativamente maiores do que o de gaveta e o de perdedor. As variáveis que discriminam entre equipes vencedoras, empate e perdedoras foram o total de chutes, chutes no gol, cruzamentos, cruzamentos contra, posse de bola e local de jogo (LAGO-PEÑAS et al., 2010).

Foi demonstrado em ambas as ligas nacionais (ARMATAS et al., 2009; KONEFAL et al., 2018; LAGO-PEÑAS et al., 2010) e campeonatos (CASTELLANO; CASAMICHANA;

LAGO, 2012; LIU et al., 2015; SZWARC, 2004) que as equipes vencedoras exibem um número total distintamente maior de arremessos e de chutes em atividade técnica de jogo alvo do que empates e perdedores. Os resultados deste estudo são semelhantes a estudos anteriores demonstrando a superioridade dos times vencedores no total de chutes sobre os critérios de alvos. Apesar de alguns resultados de pesquisas anteriores relatarem que uma proporção significativa de gols marcados em jogos de futebol de alto nível é obtida em jogos de ‘set- piece’ (cerca de 25-40%) (MITROTASIOS; ARMATAS, 2012), essa superioridade das equipes vencedoras nos critérios de chutes pode não ser observada em todos os tipos de chutes (pontapé livre, chute de fora da área de pênalti etc.) (KONEFAL et al., 2018).




Em termos gerais, as taxas de posse de bola das equipes são avaliadas de acordo com o tempo em que a bola está em jogo. Em alguns estudos (KUBAYI; LARKIN, 2020; TOKUL; MÜLAZIMOĞLU, 2018), analisando os gols marcados, eles classificam os critérios de posse de bola de acordo com o número de passes da equipe antes do gol marcado (Curto: um ou dois passes, Médio: três ou quatro passes, Longo: cinco ou mais passes). A maior porcentagem de gols marcados com posse de bola curta (50-60%) indica que ter a bola antes do gol não é uma vantagem significativa no jogo (KUBAYI; LARKIN, 2020; MITROTASIOS; ARMATAS, 2012).

Jones, James e Mellalieu (2004), times bem-sucedidos na Premier League inglesa tiveram posses significativamente mais longas do que os times malsucedidos, esta diferença também foi vista nos resultados de acordo com o status do jogo (vitória, derrota e empate). Entretanto, tanto os times bem-sucedidos quanto os malsucedidos tiveram maior duração de posse de bola quando estavam perdendo partidas, em comparação com quando estavam ganhando. No estudo atual, embora não houvesse diferença significativa, os times vencedores apresentavam uma taxa de posse de bola menor do que as outras duas situações de jogo (perda de bola). Estudos anteriores revelaram que quando os times estão perdendo, elas tendem a passar mais do que os resultados das partidas (LAGO-PEÑAS; DELLAL, 2010; LAGO, 2009). Os pesquisadores explicam que quando as equipes assumem a liderança na partida, eles preferem a defesa da equipe e jogam rápido e direto no restante do jogo. Harrop e Nevill (2014) analisaram os jogos da Liga Inglesa One. Eles descobriram que a média total de passes em jogos perdidos era significativamente maior do que em jogos empatados e ganhadores. Uma porcentagem significativamente menor de passes bem-sucedidos foi completada quando a equipe empatou. Os resultados deste estudo concluíram que as equipes deveriam fazer menos passes e dribles, mas completar mais passes e chutes com sucesso para serem bem- sucedidos. No estudo atual, que apoia estudos anteriores, quando o sucesso do time é avaliado como a classificação da liga em vez da saída da partida, os times de primeira linha têm eventos técnicos significativamente mais altos do que as classificações média e baixa. Os resultados da pesquisa de Harrop e Nevill (2014) destacaram que as variáveis 'passes', 'passes bem-sucedidos (%)', 'total de tacadas', 'dribles' e 'local da partida' (home-away) foram fatores significativos na previsão do sucesso do time. Além disso, quanto menos passes e dribles a equipe fizer, maior é a probabilidade de ganhar jogos. Na maioria dos estudos anteriores, o excesso de eventos técnicos ofensivos, como total de chutes, chutes no alvo e cruzamentos, indicam vencedores (ARMATAS et al., 2009; SZWARC, 2004). Huges e Franks (2005) descobriram que equipes bem-sucedidas transformam suas atividades de posse de bola em



mais chutes no alvo. Em um resultado semelhante, Lago-Peñas et al. (LAGO-PEÑAS et al., 2010) mostraram que a eficiência de chute foi de 46,2% em ganhos, 37,5% em empates e 37,6% em perdedores, e a importância do chute.

Em estudos anteriores, onde os resultados das partidas (ganho-perda) e a avaliação do sucesso sazonal eram feitos em partidas disputadas em casa e fora, foi revelado que o sucesso nas partidas em casa era maior. Foi examinado em estudos anteriores que as equipes definiram como "vantagem em casa" o desempenho melhor que o dos jogos fora de casa (LAGO-PEÑAS et al., 2010; LAGO-PEÑAS; DELLAL, 2010; LAGO; MARTÍN, 2007;

TAYLOR et al., 2008). Sarmento et al. (2014) nos mostra a pesquisa de revisão, os eventos de bom desempenho das equipes em partidas jogadas em casa; foi relatado que elas marcaram mais gols, chutaram mais em alvos, mais bem-sucedidos, mais dribles bem-sucedidos e mais escanteios. Em estudos anteriores, onde os resultados das partidas (ganho-perda) e a avaliação do sucesso sazonal eram feitos em partidas jogadas em casa e fora, foi revelado que o sucesso nas partidas em casa era maior.

A característica que torna este estudo valioso é a avaliação simultânea de três variáveis complementares de sucesso: classificação da temporada, resultado da partida e partida de ida e volta. O estudo foi limitado ao desempenho dos times nos jogos da Super Liga Turca na temporada 2019-2020. Em estudos futuros, ele pode ser comparado com as temporadas anteriores usando variáveis similares. Os personagens do futebol do país podem ser determinados por comparações binárias ou múltiplas com as ligas de outros países. Durante a partida, os eventos técnicos podem ser examinados de acordo com as mudanças na linha de pontuação ou partes do tempo total (trimestre ou períodos etc.). O tempo para ambos os períodos pode ser avaliado de acordo com variáveis contextuais como o primeiro período de final como processos de jogo.


Conclusão


A análise das partidas da Super Liga Turca na temporada 2019-2020 mostrou que o número total de chutes, chutes no alvo e o número de passes-chave são importantes para o sucesso de uma partida (vencedor). Todos esses eventos técnicos se revelaram importantes para o sucesso das equipes de acordo com o ranking da liga. O sucesso das equipes no ranking superior proporcionou a superioridade na porcentagem de posse de bola, chute certeiro, passes bem-sucedidos e passes-chave. De acordo com os resultados da análise, o sucesso nos jogos





em casa foi mais significativo, os times exibiram eventos técnicos mais ineficazes nos jogos fora de casa.


Aplicação prática


Os analistas podem construir um modelo para seus times, tomando como exemplo esta pesquisa atual e resultados e métodos similares de estudos de generalização. Eles também podem examinar as estatísticas de seus oponentes em termos de sucesso de partida, variáveis de casa ou de fora. Os treinadores podem criar planos de treinamento e estratégias de jogos apropriados às informações obtidas a partir dos resultados desta pesquisa. Os pesquisadores podem conduzir novas pesquisas removendo variáveis e outras amostras que estejam fora dos limites do estudo atual.


REFERÊNCIAS


ALMEIDA, C. H.; FERREIRA, A. P.; VOLOSSOVITCH, A. Effects of match location, match status and quality of opposition on regaining possession in UEFA champions league. Journal of Human Kinetics, v. 41, n. 1, p. 203-214, 2014. DOI: https://doi.org/10.2478/hukin-2014-0048


ARMATAS, V. et al. Evaluation of goals scored in top ranking soccer matches: greek

“superleague” 2006-07. Serbian Journal of Sports Sciences, v. 3, n. 1, p. 39-43, 2009.


BLOOMFIELD, J. R.; POLMAN, R. C. J.; O’DONOGHUE, P. G. Effects of score-line on

team strategies in FA Premier League Soccer. Journal of Sports Sciences, v. 23, n. 2, p. 191- 192, 2005. DOI: https://core.ac.uk/download/pdf/77287447.pdf


CARLING, C.; WILLIAMS, M. A.; REILLY, T. Handbook of Soccer Match Analysis. A systemic approach to improve performance. Routledge, Abingdon, London, UK, 2005. https://doi.org/https://doi.org/10.4324/9780203448625


CASTELLANO, J.; CASAMICHANA, D.; LAGO, C. The use of match statistics that discriminate between successful and unsuccessful soccer teams. Journal of Human Kinetics, v. 31, n. 1, p. 137-147, 2012. DOI: https://doi.org/10.2478/v10078-012-0015-7


CLEMENTE, F. M. Study of successful soccer teams on fifa world cup 2010. Pamukkale Journal of Sport Sciences, v. 3, n. 3, p. 90–103, 2012.


COHEN, J. Statistical Power Analysis for the Behavioral Sciences. 2. ed. Hillsdale, NJ: Lawrence Erlbaum Associates,1988.


DURLIK, K.; BIENIEK, P. Analysis of goals and assists dıversity in English Premier League.

Journal of Health Sciences, v. 4, n. 5, p. 47-56, 2014.




HARROP, K.; NEVILL, A. Performance indicators that predict success in an English Professional League One soccer team. International Journal of Performance Analysis in Sport, v. 14, n. 3, p. 907-920, 2014. DOI: https://doi.org/10.1080/24748668.2014.11868767


HUGHES, M.; FRANKS, I. Analysis of passing sequences, shots and goals in soccer.

Journal of Sports Sciences, v. 23, n. 5, p. 509-514, 2005.


JAMES, N. Notational analysis in soccer: past, present and future. International Journal of Performance Analysis in Sport, v. 6, n. 2, p. 67-81, 2006. DOI: https://doi.org/10.1080/24748668.2006.11868373


JONES, P. D.; JAMES, N.; MELLALIEU, S. D. Possession as a performance indicator in soccer. International Journal of Performance Analysis in Sport, v. 4, n. 1, p. 98-102, 2004. DOI: https://doi.org/10.1080/24748668.2004.11868295


KITE, C. S.; NEVILL, A. The Predictors and Determinants of Inter-Seasonal Success in a Professional Soccer Team. Journal of Human Kinetics, v. 58, n. 1, p. 157-167, 2017. DOI: https://doi.org/10.1515/hukin-2017-0084


KONEFAL, M. et al. Match outcome vs match status and frequency of selected technical activities of soccer players during UEFA EURO 2016. International Journal of Performance Analysis in Sport, v. 18, n. 4, p. 568-581, 2018. DOI: https://doi.org/10.1080/24748668.2018.1501991


KUBAYI, A.; LARKIN, P. Technical performance of soccer teams according to match outcome at the 2019 FIFA Women’s World Cup. International Journal of Performance Analysis in Sport, v. 20, n. 5, p. 908-916, 2020. DOI: https://doi.org/10.1080/24748668.2020.1809320


LAGO, C. The influence of match location, quality of opposition, and match status on possession strategies in professional association football. Journal of Sports Sciences, v. 27, n. 13, p. 1463-1469, 2009. DOI: https://doi.org/10.1080/02640410903131681


LAGO, C.; MARTÍN, R. Determinants of possession of the ball in soccer. Journal of Sports Sciences, v. 25, n. 9, p. 969-974, 2007. DOI: https://doi.org/10.1080/02640410600944626


LAGO-PEÑAS, C.; DELLAL, A. Ball possession strategies in elite soccer according to the evolution of the match-score: The influence of situational variables. Journal of Human Kinetics, v. 25, n. 1, p. 93-100, 2010. DOI: https://doi.org/10.2478/v10078-010-0036-z


LAGO- PEÑAS, C.; LAGO-BALLESTEROS, J.; REY, E. Differences in performance indicators between winning and losing teams in the UEFA Champions League. Journal of Human Kinetics, v. 27, p. 135-146, 2011. DOI: https://doi.org/10.2478/v10078-011-0011-3


LAGO-PEÑAS, C. et al. Game-related statistics that discriminated winning, drawing and losing teams from the Spanish soccer league. Journal of Sports Science and Medicine, v. 9, n. 2, p. 288-293, 2010.




LIU, H. et al, J. Match statistics related to winning in the group stage of 2014 Brazil FIFA World Cup. Journal of Sports Sciences, v. 33, n. 12, p. 1205-1213, 2015. https://doi.org/10.1080/02640414.2015.1022578


LIU, H.; HOPKINS, W. G.; GOMEZ, M. A. Modelling relationships between match events and match outcome in elite football. European Journal of Sport Science, v. 16, n. 5, p. 516- 525, 2016. DOI: https://doi.org/10.1080/17461391.2015.1042527


LIU, H. et al. Inter-operator reliability of live football match statistics from OPTA Sportsdata.

International Journal of Performance Analysis in Sport, v. 13, n. 3, p. 803-821, 2013. DOI: https://doi.org/10.1080/24748668.2013.11868690


MACKENZIE, R.; CUSHION, C. Performance analysis in football: A critical review and implications for future research. Journal of Sports Sciences, v. 31, n. 6, p. 639-676, 2013. DOI: https://doi.org/10.1080/02640414.2012.746720


MITROTASIOS, M.; ARMATAS, V. Analysis of goal scoring patterns in the 2012 European Football Championship. The Sport Journal, v. 50, p. 1-11, 2012.


NJORORAI, W. Timing of goals scored in selected European and South American soccer leagues, FIFA and UEFA tournaments and the critical phases of a match. International Journal of Sports Science, v. 4, n. 6A, p. 56-64, 2014.


REDWOOD-BROWN, A. Passing patterns before and after goal scoring in FA Premier League Soccer. International Journal of Performance Analysis in Sport, v. 8, n. 3, p. 172- 182, 2008. DOI: https://doi.org/10.1080/24748668.2008.11868458


SARMENTO, H. et al. Match analysis in football: a systematic review. Journal of Sports Sciences, v. 32, n. 20, p. 1831-1843, 2014. DOI:

https://doi.org/10.1080/02640414.2014.898852


SZWARC, A. Effectiveness of Brazilian and German Teams and the Teams Defeated By Them During the 17 Th Fifa World Cup. Kinesiology, v. 36, n. 1, p. 83-89, 2004.


TAYLOR, J. et al. The influence of match location, quality of opposition, and match status on technical performance in professional association football. Journal of Sports Sciences, v. 26, n. 9, p. 885-895, 2008. DOI: https://doi.org/10.1080/02640410701836887


TOKUL, E.; MÜLAZIMOĞLU, O. Analyzing the process from the initiation of attack to goal scoring touch in the EURO 2016. Universal Journal of Educational Research, v. 6, n. 12, p. 2738-2742, 2018. DOI: https://doi.org/10.13189/ujer.2018.061206


WHOSCORED. Fixtures/Türkiye-Super-Lig-2019-2020. 2020. Available: https://tr.whoscored.com. Access: 10 Dec. 2020.




Como referenciar este artigo


MULAZIMOGLU, O. O efeito de eventos técnicos especiais no jogo sobre o sucesso dos times de futebol profissional: Super Liga Turca. Revista online de Política e Gestão Educacional, Araraquara, v. 26, n. 2, p. 1415-1429, maio/ago. 2021. e-ISSN: 1519-9029. DOI: https://doi.org/10.22633/rpge.v25i2.15494


Submetido em: 10/05/2021

Revisões requeridas em: 25/06/2021 Aprovado em: 20/07/2021 Publicado em: 01/08/2020




THE EFFECT OF SPECIAL TECHNICAL EVENTS IN THE GAME ON THE SUCCESS OF PROFESSIONAL SOCCER TEAMS: TURKISH SUPER LEAGUE


O EFEITO DE EVENTOS TÉCNICOS ESPECIAIS NO JOGO SOBRE O SUCESSO DOS TIMES DE FUTEBOL PROFISSIONAL: SUPER LIGA TURCA


EL EFECTO DE LOS EVENTOS TÉCNICOS ESPECIALES DEL JUEGO EN EL ÉXITO DE LOS EQUIPOS DE FÚTBOL PROFESIONALES: SUPER LIGA TURCA


Olcay MULAZIMOGLU1


ABSTRACT: The purpose of this research is to examine the effect of possession, passing and shooting criteria on the success of the teams in The Turkish Soccer Super League matches in the 2019-2020 season. 612 data of each team of 306 matches played by 18 teams throughout the season were examined. In all matches, the frequencies of the teams' total number of shots, number of shots on target, percentage of possession, percentage of successful passes, number of passing, number of successful passes, key passes and rating scores criteria were collected from Whoscored’s webpage. One-way Anova analysis according to the match results (lose- draw-win) showed that means of the shots on target (p<.001), total number of shots (p=.03), key passes (p=.03) and rating score (p<.001) were significantly higher in win. In the analyzes according to the league ranking (rank 1-6, rank 7-12, rank 13-18).


KEYWORDS: Match analysis. Shots on target. Ball possession.


RESUMO: O objetivo desta pesquisa é examinar o efeito dos critérios de posse, passe e chute no sucesso dos times nos jogos da Super Liga de Futebol da Turquia na temporada 2019-2020. Foram examinados 612 dados de cada equipe de 306 partidas disputadas por 18 equipes ao longo da temporada. Em todas as partidas, as frequências do número total de chutes dos times, número de chutes no alvo, porcentagem de posse de bola, porcentagem de passes bem-sucedidos, número de passes, número de passes bem-sucedidos, passes importantes e critérios de pontuação foram coletados na página Whoscored. A análise da Anova unilateral de acordo com os resultados da partida (perder-empatar-vencer) mostrou que a média dos chutes no alvo (p <0,001), número total de chutes (p = 0,03), passes importantes (p = 0,03) e pontuação de classificação (p <0,001) foram significativamente maiores na vitória. Nas análises de acordo com a classificação da liga (classificação 1-6, classificação 7-12, classificação 13-18).


PALAVRAS-CHAVE: Análise de jogo. Chutes no alvo. Posse de bola.


RESUMEN: El propósito de esta investigación es examinar el efecto de los criterios de posesión, pase y tiro en el éxito de los equipos en los partidos de la Superliga de fútbol de Turquía en la temporada 2019-2020. Se examinaron 612 datos de cada equipo de 306


1 Mugla Sitki Kocman University (MSKU), Muğla – Turkey. Associate Professor of the Faculty of Sports Sciences. ORCID: https://orcid.org/0000-0001-5599-280X. Email: olcaymulazimoglu@mu.edu.tr

RPGE– Revista on line de Política e Gestão Educacional, Araraquara, v. 25, n. 2, p. 1418-1431, May/Aug. 2021. e-ISSN: 1519-9029



partidos jugados por 18 equipos a lo largo de la temporada. En todos los partidos, las frecuencias del número total de tiros de los equipos, el número de tiros a puerta, el porcentaje de posesión, el porcentaje de pases exitosos, el número de pases, el número de pases exitosos, los pases clave y los criterios de puntuación de calificación se recopilaron de la página web de Whoscored. El análisis de Anova unidireccional de acuerdo con los resultados del partido (perder-empatar-ganar) mostró que las medias de los tiros a puerta (p

<.001), el número total de tiros (p = .03), pases clave (p = .03) y el puntaje de calificación (p

<.001) fueron significativamente más altos en win. En los análisis según la clasificación de la liga (rango 1-6, rango 7-12, rango 13-18).


PALABRAS CLAVE: Análisis del partido. Tiros a puerta. Posesión del balón.


Introduction


One of the most important problems of coaches and analysts in football teams is to obtain video analysis and statistics of their own teams and opponents. The notational analysis of soccer matches is helpful in optimizing feedback for coaches and players aiming at performance improvement (CARLING; WILLIAMS; REILLY, 2005).

In the field of football performance analysis, previously recorded match videos were widely used to monitor, evaluate, and analyze team performances (JAMES, 2006; LIU; HOPKINS; GOMEZ, 2016), but more recently modern video analysis systems such as AMISCO, OPTA and ProZone provide a comprehensive database of soccer games (LIU et al., 2013; LIU; HOPKINS; GOMEZ, 2016; MACKENZIE; CUSHION, 2013).

Player performance analyzes associated with match results can be very useful in explaining the effects on team performance (LAGO, 2009; SARMENTO et al., 2014). Many studies based on notational match analysis have focused on the number of match outcome (win, draw, loss) and technical activities performed by the players (CASTELLANO; CASAMICHANA; LAGO, 2012; CLEMENTE, 2012; KITE; NEVILL, 2017; LAGO- PEÑAS et al., 2010; LIU et al., 2015).

Studies have been carried out to associate technical and tactical events in matches played in various championships at world and continental level (World Cup, European Championship, etc.) and football leagues with team success. In addition, studies examining the goals scored in championships according to various technical and tactical criteria evaluated success based on goals (DURLIK; BIENIEK, 2014; NJORORAI, 2014; TOKUL; MÜLAZIMOĞLU, 2018).

Studies examining the technical efficiency of the players according to the score-line (winning, draw, losing) changes during the match and the match result have revealed the




effect of the season-end success (ALMEIDA; FERREIRA; VOLOSSOVITCH, 2014; LAGO- PEÑAS; DELLAL, 2010; REDWOOD-BROWN, 2008).

Some research assessed the frequency of performance of selected technical activities in relation to match outcome and match status in matches of the championships. The most important are frequencies of different types of passes and higher percent of ball possession when the team is playing to change an unfavorable score (losing and drawing) (KONEFAL et al., 2018).

In a study conducted in the Spanish league, they reported that home teams and losing teams that conceded goals had more ball possession. The opponent's identity, the worse the opponent, the greater the possession of the ball (BLOOMFIELD; POLMAN; O’DONOGHUE, 2005; LAGO; MARTÍN, 2007).

In study of the 1990 and 1994 World Cup by Hughes and Franks (2005) was identified that successful teams produced more goals per possession using longer passing sequences. However unsuccessful teams were found to adopt a more direct style of play, using shorter passing sequences, as these teams didn't have a sufficient level of skill to sustain possession of the ball. (HUGHES; FRANKS, 2005).

Many other research studies have also supported the idea that styles of play differ between successful and unsuccessful teams when a variety of different performance indicators have been analyzed (LAGO-PEÑAS et al., 2011; LAGO-PEÑAS et al., 2010).

In addition to these researches, football viewers, fans, TV programmers are their approaches to game analysis results. Some statistics are reflected on the screen while the matches are played. Basic information such as the percentage of teams' possession of the ball, total pass / on-target pass, total shots / shots on target are displayed on the screen. Football commentators evaluate the superiority of the teams in their comments while watching the match summaries on TV programs by also mentioning these statistics. An important issue to be wondered about; Is more Ball Possession and more passing a sign of superiority in the game? Will shooting more than your opponents bring your team the championship at the end of the league?

The purpose of this research is to examine the effect of possession, passing and shooting criteria on the success of the teams in winning games and league rankings in The Turkish Football Association (TFA) Super League matches in the 2019-2020 season.




Methods


Sample, data resource and variables


The Turkish Football Association (TFA) Super League is the highest level of professional soccer in Turkey which into the top leagues in Europe. Each season begins in August (late summer in Turkey) and in May (late spring) ends. In each season, every team plays against each of the other teams twice, once at home and the other away. In the 2019- 2020 season league system, with 18 teams in the League, each team plays 34 games for a total of 306 games in the season. The end-of-season rank is determined by the points (win for 3 points, draw for 1, loss for 0) summed from the 30 games of each team. All the 306 matches in the TFA Super League of the 2019-2020 season were chosen as the sample of the current study. 34 matches played by each of the 18 teams in the league were evaluated as separate participants, and variables belonging to a total of 612 participants formed the data set. Parameters obtained from each game were compared according to their success.

At the end of the season in the Super League top five teams from Turkey are entitled to participate in the UEFA League. However, one of the top five teams lost their right to participate due to the UEFA penalty. He was replaced by the sixth-ranked team. On the other hand, although the leagues were suspended this season due to the Covid-19 outbreak, all matches were played. In order to prevent the negative impact of the pandemic on the clubs, the three teams that should be relegated by the Federation decision were not relegated.

Performance-related data of these matches were collected from public accessed websites “https://tr.whoscored.com” whose data resources is “Whoscored” supported by OPTA Sportsdata Company (WHOSCORED, 2020).

The inter-operator reliability of the company’s tracking system (OPTA Client System) used to collect football match statistics was identified to be on an acceptable level. Furthermore, reliability of the tracking system (OPTA Client System) used by OPTA Sportsdata Company to collect soccer match statistics has been tested by Liu and colleagues (LIU et al., 2013), which showed that teams match events coded by independent operators using this system reached a very good agreement (weighted kappa values were .92 and .94) (LIU et al., 2013). The study was conducted in compliance with the Declaration of Helsinki. Ethics committee approval of the current study was gained from the Muğla Sitki Koçman University (Application No: 200346). Permission to use the statistics published on the "whoscored" website has been obtained.





In all matches, the frequencies of the teams' total number of shots, number of shots on target, percentage of possession, percentage of successful passes, number of passing, number of successful passes, key passes and OPTA rating criteria were collected.The success of the teams was analyzed according to the match results (win-draw-lose) and their ranking (upper- level teams: 1-6 ranked “UEFA” group, mid-level teams: 7-12 ranked group, lower-level teams: 13-18 ranked relegation line and nearby teams) at the end of the season and the home- away status of the team.

Teams are divided into three groups while being evaluated according to their success order. The first group was made up of teams that finished the league in the top six. The second group was the next six teams, that is, teams from 7-12. The third group was made up of six teams at the end of the ranking.


Definition of parameters


Shot: Kicks with any legal part of the body for the purpose of scoring a goal, whether it is accurate or not.

Shot on Target: All goal attempts that could result in a goal if not blocked or saved. Passing: Balls voluntarily given by a player to a teammate

Pass success: voluntary passes taken by a player are taken by his teammate

Ball Possession: The amount of time a team controlled the ball during a game, from the moment it takes over the ball from the opposing team without any clear interruption, as a proportion of total time when the ball was in play (Konefal et al., 2018).


Statistical analysis


All variables were checked for normal distribution (A kurtosis value between was

±1.0) and homogeneity of variance (Levene’s test). Arithmetic means and standard errors were calculated. To compare the mean values of the examined variables a one-way ANOVA was used. When a significant effect size was found, a post-hoc Fisher’s LSD test was performed. The level of statistical significance was set at p < .05.

Moreover, partial eta squared (η2) was calculated, and the effect sizes were determined: small effect size (≥0.01 and <0.06), medium effect size (≥0.06 and <0.14), large effect size (≥0.14) (Cohen, 1988). All statistical analyses were made using the SPSS (version 22.0, SPSS Inc., Chicago, IL) software package.




Results


The data revealed that there was a significant difference in the mean of total shoots (p

=.03) and shoot on target (p <.001) according to the teams' winning, drawn and losing status. Also, a significant difference was found in the key pass average (p =.03) and OPTA rating (p

<.001) by game success. The multiple comparison test (LSD) revealed more total shooting, shoot on target and key passes for winner in the game than loser and tied teams. “Partial eta squared” found large in the shoot on target (Table 1).


Table 1 – Comparison analysis of variables (ANOVA) and effect size according to the game result (mean±sd)


                            Game Result                             

Variable

1 Lose

(n=224)

2 Draw

(n=164)

3 Win

(n=224)

F(Sig.)

P.eta Post Hoc

squared (LSD)

(ɳ2)

OPTA rating 6.36±.20 6.67±.12 7.02±.24 611.41(.001)** .668-L 3>1,2, 2>1


Total shot

12.46±5.15

12.57±5.36

13.67±4.90

3.73(.025)*

.012-S

3>1,2

Shot on target

3.65±2.12

4.09±2.16

5.55±2.44

43.20(.001)**

.124-L

3>1,2

Ball possession %

50.45±10.55

50.02±10.38

49.55±10.55

0.42(.659)

.001-S

null

Pass success %

79.75±5.83

78.30±6.83

79.69±6.78

2.91(.055)

.009-S

1>2, 3>2

Total pass

426.75±96.98

416.24±96.25

422.72±103.98

0.53(.589)

.002-S

null

Pass success

344.06±97.56

330.78±98.22

342.79±106.56

0.95(.388)

.003-S

null


Key pass

9.46±4.17

9.37±4.46

10.35±4.14

3.46(.032)*

.011-S

3>1,2

Significance level: * p <.05; ** p <.01

ES: effect size (ɳ2); S: small (≥.01 and <.06), M: medium (≥.06 and <.14), L: large (≥.14)

Source: Prepared by the author


The data revealed a significant difference in all variables of the teams grouped according to league rankings. The top one-third ranked teams recorded more total shots (respectively: 14.48±5.10; 12.82±5.16 11.50±4.74) and shots on target (respectively: 5.18±2.48; 4.34±2.35; 3.88±2.20) than the second and third one-third teams (p<.001). In terms of possession percentage, the first one-third teams had a higher percentage (respectively

: 53.29±10.15%; 49.22±10.34%; 47.51±10.20%: p<.001). A significant difference was found only between the first and last group in the variable of pass success percentage (respectively: 80.50±6.75%; 78.22±6.28%; p=.002). In the number of total pass, pass success and key pass, the first one-third was found to be significantly higher than the other two subgroups (p<.001). While the effect size was medium in OPTA rating and key pass variables among the success groups at the end of the league, all other variables were small (Table 2).




Table 2 – Comparison analysis of variables (ANOVA) and effect size according to the league rank (mean±sd)


                           Rank of League                            

Variable

1) Rank 1-6

(n=204)

2) Rank 7-12 3) Rank 13-18

F(Sig.)

(n=204)

(n=204)

P.eta Post Hoc

squared (LSD)

(ɳ2)

OPTA rating 6.82±.32 6.64±.35 6.58±.31 28.80(.001)* .086-M 1>2,3


Total shooting

14.48±5.10

12.82±5.16

11.50±4.74

18.23(.001)*

.056-S

1>2,3, 2>3

Shot on target

5.18±2.48

4.34±2.35

3.88±2.20

16.09(.001)*

.050-S

1>2,3, 2>3

Ball possession %

53.29±10.15

49.22±10.34

47.51±10.20

17.19(.001)*

.053-S

1>2,3

Pass success %

80.50±6.75

79.30±6.24

78.22±6.28

6.39(.002)*

.021-S

1>3

Total pass

454.01±101.26

408.64±95.14

404.73±94.26

16.30(.001)*

.051-S

1>2,3

Pass success

371.93±105.21

328.29±96.09

319.90±94.34

16.36(.001)*

.051-S

1>2,3


Key pass

11.26±4.20

9.46±4.26

8.56±3.85

22.94(.001)*

.070-M

1>2,3

Significance level: * p <.01

ES: effect size (ɳ2); S: small (≥.01 and <.06), M: medium (≥.06 and <.14), L: large (≥.14)

Source: Prepared by the author


Independent sample t-test results revealed a significant difference between the teams' home and away matches in all game variables. The home performances of the teams for all variables were higher than away. Teams played with more ball possession (respectively: 51.4±10.41%; 48.61±10.41; p<.001) in home matches than away, and played with a higher percentage of pass success (respectively: 80.09±6.28; 78.60±6.61; p=.004). A significantly high average total shot was found at home, and the average shot on target was also higher than outside. The key pass average was higher in home games. OPTA's rating also scored higher points in home matches (Table 3).


Table 3 – Comparison analysis of variables (independent samples t-test) and effect size according to the home and away team


Variable

Home

(n=306)

Away

(n=306)

t (Sig.)

ɳ2

OPTA rating 6.75±.34 6.62±.34 4.81(.001)** .037-S

Total shooting

14.13±5.14

11.73±4.87

5.94(.001)**

.055-S

Shot on target

4.9±2.46

4.03±2.26

4.55(.001)**

.033-S

Ball possession %

51.4±10.41

48.61±10.41

3.32(.001)**

.018-S

Pass success %

80.09±6.28

78.60±6.61

2.86(.004)**

.013-S

Total pass

432.9±100.00

412.10±97.72

2.60(.01)*

.011-S

Pass success

351.80±101.17

328.30±99.85

2.89(.004)**

.013-S

Key pass

10.62±4.24

8,90±4,09

5.12(.001)**

.041-S

Significance level: * p <.05; ** p <.01

ES: effect size (ɳ2); S: small (≥.01 and <.06), M: medium (≥.06 and <.14), L: large (≥.14)

Source: Prepared by the author




Discussion


We often talk about football or criticize the game with your friends about the game after a match in social life. First of all, we evaluate the superiority of the teams in terms of the goals scored. The goal positions of the teams, shots and set play are important game events that are criticized. The most important scientific evidence you use against your opponent in this match criticism discussion will be the match statistics such as the teams' possession rate, the total number of passes and the correct pass percentage, the number of shots on the goal. Could the game statistics really tell us that for a football match one team is superior to another? Could the statistics of a match or all matches in a season clearly reveal the success of the team? This study produced results that answered questions that people often use seeking the truth of this information.

According to this study' results, in the 2019-2020 season of the Turkish Super League, the most important distinctive game event in the success of the teams that won the match was the number of shots and shooting accuracy variables. Significantly higher number of key passes was the variable explaining more shot averages. Eight game events examined in this study revealed that sustained seasonal success is more important than a game success. Contrary to the seasonal success (league rank), it was an important result of this study that there was no difference between the winning, losing and tied teams in the game both ball possession and percentage of successful pass variables.

Konefal et al. (2018) examined the performance frequency of selected technical activities by soccer teams playing in the EURO 2016 (only group stage) in relation to match outcome and match status. This study revealed that the most important technical activity for the teams at the end of a match is the number of shots on the goal and the shot success rate. However, when the team is playing to change an unfavorable score (losing and drawing in match status), the most important are frequencies of different types of passes and higher percent of ball possession (KONEFAL et al., 2018).

Lago-Peñas et al. (2018) analyzed various technical activities in men’s professional soccer league matches. Winning teams had averages for total shots (W: 14.4±5.1, D: 13.6±5.2, L: 11.9±4.8), shots on goal (W: 6.6±2.8, D: 5.1±2.7, L: 4.2±2.4) that were

significantly higher than drawer and loser. The variables that discriminate between winning, drawing and losing teams were the total shots, shots on goal, crosses, crosses against, ball possession and venue (LAGO-PEÑAS et al., 2010).




It has been demonstrated in both national leagues (ARMATAS et al., 2009; KONEFAL et al., 2018; LAGO-PEÑAS et al., 2010) and championships (CASTELLANO; CASAMICHANA; LAGO, 2012; LIU et al., 2015; SZWARC, 2004) that winning teams exhibit a distinctively higher total number of shots and shots on target play technical activity than draws and losers. The results of this study are similar to previous studies demonstrating the superiority of winning teams in the total shot and shot on target criteria. Despite some previous research results reporting that a significant proportion of goals scored in high-level football matches are scored from set-piece play (about 25--40%) (MITROTASIOS; ARMATAS, 2012), this superiority of winning teams in the shooting criteria may not be observed in all shot types (free kick, shooting from outside the penalty area, etc.) (KONEFAL et al., 2018).

In general terms, the rates of possession of the teams are evaluated according to the time the ball is in play. In some studies (KUBAYI; LARKIN, 2020; TOKUL; MÜLAZIMOĞLU, 2018), analyzing the goals scored, they classified the criteria of possession according to the number of passes of the team before the goal scored (Short: one or two passes, Medium: three or four passes, Long: five or more passes). The higher percentage of goals scored with short ball possession (50-60%) indicates that having the ball before the goal is not a distinctive significant game advantage (KUBAYI; LARKIN, 2020; MITROTASIOS; ARMATAS, 2012).

Jones, James and Mellalieu (2004) successful teams in the English Premier League were found to have significantly longer possessions than unsuccessful teams, this difference was also seen in the results according to the match status (winning, losing, and drawing). However, both successful and unsuccessful teams had longer durations of possession when they were losing matches compared to when winning. In the current study, though there was no significant difference, the winning teams showed a lower ball possession rate than the other two match situations (lose-draw). Previous studies have revealed that when teams are losing, they tend to pass more than match outcomes (LAGO-PEÑAS; DELLAL, 2010; LAGO, 2009). The researchers explain that when teams took the lead in the match, they preferred team defense and played fast and direct in the remainder of the game. Harrop and Nevill (HARROP; NEVILL, 2014) analyzed English League One matches. They found that the total average pass in lost matches was significantly higher than in tied and winning matches. A significantly lower percentage of successful passes were completed when the team drew. This study results concluded that the teams should perform fewer passes and dribbles but complete more successful passes and shots to be successful. In the current study, which

RPGE– Revista on line de Política e Gestão Educacional, Araraquara, v. 25, n. 2, p. 1418-1431, May/Aug. 2021. e-ISSN: 1519-9029



supports previous studies, when the team success is evaluated as the league rank instead of the match output, the top-ranking teams have significantly higher technical events than the middle and lower ranks. Harrop and Nevill's research results highlighted that the variables 'passes', 'successful passes (%)', 'total shots', 'dribbles' and 'match location (home-away)' were significant factors in predicting the team's success. Besides, the fewer passes and dribbles the team perform, the more likely they are of winning games. In most previous studies, the excess of offensive technical events such as total shots, shots on target and crosses indicate winners (ARMATAS et al., 2009; SZWARC, 2004). Huges and Franks (2005) found that successful teams turn their possession activities into more shooting on target. In a similar result, Lago- Peñas et al (LAGO-PEÑAS et al., 2010) showed that shooting efficiency was 46.2% in gains, 37.5% in draws and 37.6% in losers, and the importance of shooting.

In previous studies where match outcomes (win-draw-lose) and seasonal success evaluation were made in matches played at home and away, it was revealed that the success in home matches was higher. It was examined in previous studies that the teams defined as "home advantage" performed better than the away games in the matches they played at home (LAGO-PEÑAS et al., 2010; LAGO-PEÑAS; DELLAL, 2010; LAGO; MARTÍN, 2007;

TAYLOR et al., 2008). Sarmento et al. (2014)’ the review research, the good performance events of the teams in matches played at home; it was reported that they scored more goals, shot more targets, more passes and successful passes, more successful dribbles and more corners. In previous studies where match outcomes (win-draw-lose) and seasonal success evaluation were made in matches played at home and away, it was revealed that the success in home matches was higher.

The feature that makes this study valuable is the simultaneous evaluation of three complementary success variables: season ranking, match outcome and home-away match. The study was limited to the performances of the teams in the matches in the Turkish Super League in the 2019-2020 season. In future studies, it can be compared with previous seasons using similar variables. Country football characters can be determined by binary or multiple comparisons with other country leagues. During the match, technical events can be examined according to the score-line changes or parts of the total time (quarter time or periods etc.). The time for both periods can be evaluated according to contextual variables as the first-middle- end period as game processes.





Conclusion


The analysis of the matches in the Turkish Super League in the 2019-2020 season showed that the total number of shots, shots on target and the number of key passes were an important determinant of a match success (winner). All these technical events turned out to be important in the success of the teams according to the league rank. The success of the teams in the top rankings provided the superiority in the percentage of possession, accurate shooting, successful passes, and key passes. According to the analysis results, the success in the home matches was more meaningful, the teams exhibited more ineffective technical events in the away matches.


Practical application


Analysts can build a model for their teams by taking this current research and similar generalization study results and methods as an example. They can also examine their opponents' statistics in terms of match success, home, or away variables. Coaches can create training plans and match strategies appropriate to the information obtained from this research results. Researchers can conduct new research by removing variables and other samples that are outside the current study boundaries.


REFERENCES


ALMEIDA, C. H.; FERREIRA, A. P.; VOLOSSOVITCH, A. Effects of match location, match status and quality of opposition on regaining possession in UEFA champions league. Journal of Human Kinetics, v. 41, n. 1, p. 203-214, 2014. DOI: https://doi.org/10.2478/hukin-2014-0048


ARMATAS, V. et al. Evaluation of goals scored in top ranking soccer matches: greek

“superleague” 2006-07. Serbian Journal of Sports Sciences, v. 3, n. 1, p. 39-43, 2009.


BLOOMFIELD, J. R.; POLMAN, R. C. J.; O’DONOGHUE, P. G. Effects of score-line on

team strategies in FA Premier League Soccer. Journal of Sports Sciences, v. 23, n. 2, p. 191-192, 2005. DOI: https://core.ac.uk/download/pdf/77287447.pdf


CARLING, C.; WILLIAMS, M. A.; REILLY, T. Handbook of Soccer Match Analysis. A systemic approach to improve performance. Routledge, Abingdon, London, UK, 2005. https://doi.org/https://doi.org/10.4324/9780203448625


CASTELLANO, J.; CASAMICHANA, D.; LAGO, C. The use of match statistics that discriminate between successful and unsuccessful soccer teams. Journal of Human Kinetics, v. 31, n. 1, p. 137-147, 2012. DOI: https://doi.org/10.2478/v10078-012-0015-7



CLEMENTE, F. M. Study of successful soccer teams on fifa world cup 2010. Pamukkale Journal of Sport Sciences, v. 3, n. 3, p. 90–103, 2012.


COHEN, J. Statistical Power Analysis for the Behavioral Sciences. 2. ed. Hillsdale, NJ: Lawrence Erlbaum Associates,1988.


DURLIK, K.; BIENIEK, P. Analysis of goals and assists dıversity in English Premier League. Journal of Health Sciences, v. 4, n. 5, p. 47-56, 2014.


HARROP, K.; NEVILL, A. Performance indicators that predict success in an English Professional League One soccer team. International Journal of Performance Analysis in Sport, v. 14, n. 3, p. 907-920, 2014. DOI: https://doi.org/10.1080/24748668.2014.11868767


HUGHES, M.; FRANKS, I. Analysis of passing sequences, shots and goals in soccer.

Journal of Sports Sciences, v. 23, n. 5, p. 509-514, 2005.


JAMES, N. Notational analysis in soccer: past, present and future. International Journal of Performance Analysis in Sport, v. 6, n. 2, p. 67-81, 2006. DOI: https://doi.org/10.1080/24748668.2006.11868373


JONES, P. D.; JAMES, N.; MELLALIEU, S. D. Possession as a performance indicator in soccer. International Journal of Performance Analysis in Sport, v. 4, n. 1, p. 98-102, 2004. DOI: https://doi.org/10.1080/24748668.2004.11868295


KITE, C. S.; NEVILL, A. The Predictors and Determinants of Inter-Seasonal Success in a Professional Soccer Team. Journal of Human Kinetics, v. 58, n. 1, p. 157-167, 2017. DOI: https://doi.org/10.1515/hukin-2017-0084


KONEFAL, M. et al. Match outcome vs match status and frequency of selected technical activities of soccer players during UEFA EURO 2016. International Journal of Performance Analysis in Sport, v. 18, n. 4, p. 568-581, 2018. DOI: https://doi.org/10.1080/24748668.2018.1501991


KUBAYI, A.; LARKIN, P. Technical performance of soccer teams according to match outcome at the 2019 FIFA Women’s World Cup. International Journal of Performance Analysis in Sport, v. 20, n. 5, p. 908-916, 2020. DOI: https://doi.org/10.1080/24748668.2020.1809320


LAGO, C. The influence of match location, quality of opposition, and match status on possession strategies in professional association football. Journal of Sports Sciences, v. 27, n. 13, p. 1463-1469, 2009. DOI: https://doi.org/10.1080/02640410903131681


LAGO, C.; MARTÍN, R. Determinants of possession of the ball in soccer. Journal of Sports Sciences, v. 25, n. 9, p. 969-974, 2007. DOI: https://doi.org/10.1080/02640410600944626


LAGO-PEÑAS, C.; DELLAL, A. Ball possession strategies in elite soccer according to the evolution of the match-score: The influence of situational variables. Journal of Human Kinetics, v. 25, n. 1, p. 93-100, 2010. DOI: https://doi.org/10.2478/v10078-010-0036-z





LAGO- PEÑAS, C.; LAGO-BALLESTEROS, J.; REY, E. Differences in performance indicators between winning and losing teams in the UEFA Champions League. Journal of Human Kinetics, v. 27, p. 135-146, 2011. DOI: https://doi.org/10.2478/v10078-011-0011-3


LAGO-PEÑAS, C. et al. Game-related statistics that discriminated winning, drawing and losing teams from the Spanish soccer league. Journal of Sports Science and Medicine, v. 9, n. 2, p. 288-293, 2010.


LIU, H. et al, J. Match statistics related to winning in the group stage of 2014 Brazil FIFA World Cup. Journal of Sports Sciences, v. 33, n. 12, p. 1205-1213, 2015. https://doi.org/10.1080/02640414.2015.1022578


LIU, H.; HOPKINS, W. G.; GOMEZ, M. A. Modelling relationships between match events and match outcome in elite football. European Journal of Sport Science, v. 16, n. 5, p.

516-525, 2016. DOI: https://doi.org/10.1080/17461391.2015.1042527


LIU, H. et al. Inter-operator reliability of live football match statistics from OPTA Sportsdata. International Journal of Performance Analysis in Sport, v. 13, n. 3, p. 803- 821, 2013. DOI: https://doi.org/10.1080/24748668.2013.11868690


MACKENZIE, R.; CUSHION, C. Performance analysis in football: A critical review and implications for future research. Journal of Sports Sciences, v. 31, n. 6, p. 639-676, 2013. DOI: https://doi.org/10.1080/02640414.2012.746720


MITROTASIOS, M.; ARMATAS, V. Analysis of goal scoring patterns in the 2012 European Football Championship. The Sport Journal, v. 50, p. 1-11, 2012.


NJORORAI, W. Timing of goals scored in selected European and South American soccer leagues, FIFA and UEFA tournaments and the critical phases of a match. International Journal of Sports Science, v. 4, n. 6A, p. 56-64, 2014.


REDWOOD-BROWN, A. Passing patterns before and after goal scoring in FA Premier League Soccer. International Journal of Performance Analysis in Sport, v. 8, n. 3, p. 172-182, 2008. DOI: https://doi.org/10.1080/24748668.2008.11868458


SARMENTO, H. et al. Match analysis in football: a systematic review. Journal of Sports Sciences, v. 32, n. 20, p. 1831-1843, 2014. DOI:

https://doi.org/10.1080/02640414.2014.898852


SZWARC, A. Effectiveness of Brazilian and German Teams and the Teams Defeated By Them During the 17 Th Fifa World Cup. Kinesiology, v. 36, n. 1, p. 83-89, 2004.


TAYLOR, J. et al. The influence of match location, quality of opposition, and match status on technical performance in professional association football. Journal of Sports Sciences, v. 26, n. 9, p. 885-895, 2008. DOI: https://doi.org/10.1080/02640410701836887


TOKUL, E.; MÜLAZIMOĞLU, O. Analyzing the process from the initiation of attack to goal scoring touch in the EURO 2016. Universal Journal of Educational Research, v. 6, n. 12, p. 2738-2742, 2018. DOI: https://doi.org/10.13189/ujer.2018.061206




WHOSCORED. Fixtures/Türkiye-Super-Lig-2019-2020. 2020. Available: https://tr.whoscored.com. Access: 10 Dec. 2020.


How to reference this article


MULAZIMOGLU, O. The effect of special technical events in the game on the success of professional soccer teams: Turkish Super League. Revista online de Política e Gestão Educacional, Araraquara, v. 25, n. 2, p. 1418-1431, maio/ago. 2021. e-ISSN: 1519-9029. DOI: https://doi.org/10.22633/rpge.v25i2.15494


Submitted: 10/05/2021 Required revisions: 25/06/2021 Approved: 20/07/2021 Published: 01/08/2021