I've just started recording stats for USWNT games and I am having trouble determining what exactly is considered a Shot On Goal and a also Shot in certain situations/circumstances. Everywhere I've searched gives me the exact same definition of a SoG... "A shot on goal is a shot that is on net. The results of a shot on goal must be either a save by the goalkeeper or defending team or a goal by the attacking team. A shot that hits the post or crossbar without being deflected by a goalkeeper or defender and does not cross the goal line is not a shot on goal." A shot on net and a shot that is saved by the keeper are easy to capture. My problem is when the shot is blocked or deflected by a defender. Is it a Shot and a Shot On Goal if a player shoots from 18 yards out and it is blocked immediately or a couple yards away from the shooter and still 16+ yards out? Is there some kind of distance that is required? What about a free kick from just outside the penalty box and the shooter shoots the ball directly into the wall that is 10 yards away? It was intended to be a shot and would be on goal if no one was in the way. Is that a Shot On Goal if they just drill it into the wall?
If a shot is blocked by a player other than the goalkeeper, it isn't a shot on goal. I agree with the definition you posted in red, but I've seen game statistics that didn't match that. Note that an own goal does not count as a shot on goal. The New York Red Bulls had a game where they allowed 5 shots on goal, 5 goals, had 1 save, and scored 1 own goal.
Thanks for the response. I think that rule contradicts what you said if I'm reading your response correctly. The definition in red says that a shot blocked by a defender (other than the goalkeeper) can also be considered a shot on goal. "The results of a shot on goal must be either a save by the goalkeeper OR defending team" I'm not bringing this up to argue with you. I'm just trying to figure this out. It takes me about an hour to go through each game and record all the stats (I'm recording about 30 different stats) and I don't want to have to go back through and watch every shot taken again. Right now, this stat seems very arbitrary to me.
http://www.flashscore.com/match/OAcT0ADJ/#match-statistics;0 is about yesterday's Liverpool-Manchester United game. Note that for each club, "Goal Attempts" = "Shots on Goal" + "Shots off Goal" + "Blocked Shots," where "Blocked Shots" means it was blocked by the opposing club. If some shots counted as "Shots on Goal" and "Blocked Shots," the right side of the equation would be higher than the left side. Also note that each club's "Shots on Goal" = their Goals + "Saves" by the other club. There were no own goals in that game. It may be possible to find another website with different statistics for the game (I once looked at possession percentages and found significant differences between sources), but I like the way Flash Score does it. http://www.premierleague.com/en-gb/...016/epl.match-stats.html/liverpool-vs-man-utd shows how the Premier League does it. There "Total Shots" = "Shots on Target" + "Shots off Target," with "Blocks" listed separately and blocked shots not counting towards "Total Shots." The Premier League and Flash Score agree about how many shots on target, shots off target, and blocked shots each club had except for how many Liverpool shots were blocked by Manchester United, where the Premier League says 4 and Flash Score says 5. The NHL doesn't count blocked shots as Shots on Goal, and I don't think soccer should either.
Thanks, I like the way Flash Score records "Goal Attempts" = "Shots on Goal" + "Shots off Goal" + "Blocked Shots" and am going to use that. This was a lot of help!!!
There's, I think, a difference between a shot where the block is a effectively a "save", like clearing from off or near the line, and a block from close range, which means the shot gets deflected away almost immediately after being hit. That said, it's something of a peeve of mine that these days people will try to use shots on target as some kind of meaningful measure of how many good scoring chances one team has had, with it being implied that a shot on target is a better chance than one off target, so a team that has 8 on target and 5 off had created better chances than one that had 5 on target and 8 off. A penalty miss that goes wide, or an open goal miss, are both off target, for example.
A shot a target is better-executed than a shot off target, but I agree that knowing whether a shot was on target or off target doesn't tell you the probability of a goal being scored from there before the shot happened. I understand how you feel, but shots on target, shots off target, and blocked shots should be easy to count. With a large sample of games and advanced computing, maybe you could divide the field into very small squares and measure the percentage of shots from each square area that went in the goal with shots being broken down into run of play, free kick, penalty kick, and maybe corner kick (goals directly from corner kicks are rare but possible). Then if a team had five shots with probabilities of a goal of 0.5, 0.4, 0.3, 0.2, and 0.1, you could say the team had an expected 1.5 goals based on the shots and their locations.
Oh, I agree, it's much easier to count. And when used in cumulative stats over a number of games, it does give a picture of shooting accuracy. My problem is that they are a crude and ineffective way of measuring anything really meaningful in individual games. A hugely lopsided tally in one team's favour would indicate domination, but the total shots tally would do that just as well. My problem is when people start trying to judge how well two teams compared played based on the shots on target stat, as it's often a poor indication - as are possession stats. Stats that just record a basic count of what happened are next to useless, as you need stuff to indicate why things happened. A few decades ago Aussie Rules football had a similar issue with stats, and they sat down and worked out measures that really made a difference. Football needs to do a similar thing.