Forecasting the results of football matches on the Internet based information

Abstract

The purpose of the article is making a model of results forecasting for football matches, which works better than bookmakers organizations. Lately the popularity of football forecasting has been increased. The existing statistic approaches show difficult and low prediction. The developed model for predicting the results of football matches uses information about the previous results of the teams. The forecast is based on forecasting factors. Although it is difficult to consider all the factors that influence the results of matches the model makes an attempt to find the most important ones. The described model of forecasting takes into consideration such characteristics as: place in the tournament table; total points; goal difference; total number of players, skipping the match; points home / away; home match; "defence strength"; "attack strength"; team form. Testing shows that forecasting and actual results of football matches coincide. The offered model could be used in commercial computer programs for forecasting results of football matches in bookmakers organizations.

Статья посвящена модели прогнозирования результатов футбольных матчей, превосходящая шансы букмекеров. Статистические подходы показали сложные и низкие результаты прогнозирования. Инструменты интеллектуального анализа данных с недостаточными возможностями, также дали низкие прогнозы. Предложенная модель использует информацию о предыдущих результатах команды. Прогноз основан на прогнозирующих факторах. Хотя трудно учесть все факторы, влияющие на результаты матчей, делается попытка найти наиболее значимые факторы.