clock menu more-arrow no yes mobile

Filed under:

Joe Hart: Career Save Percentage

I'm going to take a look at everyone's favourite topic: Joe Hart. Now, I could fill this article with my take on the merits of dropping the player from the first team, but I won't. Joe was dropped from the Man City team because his performances were beginning to hinder the team in it's ability to win football matches. For me, it's that simple.

Follow Us

Correct me if I am wrong but I believe there may be three or four trains of thought concerning Joe Hart:

  1. Joe Hart is an excellent 'keeper suffering through a horrendous patch of form.
  2. Joe Hart was was an excellent 'keeper who has suffered some irreparable decline in ability in the last year or so.
  3. Joe Hart has never been a good 'keeper.
  4. Joe Hart's drop off in form is due to his arrogance. (it's the media who laughably push this line)

I'm not sure all of the above points are relevant let alone true. But, hey. Options 3 and 4 can probably be discounted from the list above due to their factual inaccuracy. Although I could be convinced on #3 if someone can show me evidence that Hart has been sub-par by some baseline or other for his entire career.

So that leaves us with two options: Hart is broken temporarily or Hart is broken permanently. Or, maybe Hart's poor performances this season are something that 'keepers suffer through - variance, bad luck, concentration issues, personal problems or whatever.

To try and discover the issues with his current form I am going to look at Hart's City career numbers - do they show a particular trend in performance; is Hart's start to the season a rare run of bad form when placed against the rest of his career. To try and discover some of these things I am going to have to use something I am not entirely happy with and that something is called Save Percentage.

Save percentage is goals conceded/shots on target faced. The league wide average, which is more or less stable over a number of seasons, is ~69%.

The good thing about Save% is that it tells us the answer to a simple question: what percentage of shots on target did 'keeper X stop? The bad thing about save% is that the samples are tiny tiny small from game to game. It's a given that we don't know where opponents shots on target originate from or how close to goal those shots on target were shot from.

So there are problems with the numbers I am using, but I don't have anything else, bar my stupid opinion, to use and examine.

Joe Hart's PL Save%: 2010-2013

Firstly, Hart's save% on an individual game basis and a rolling basis.


The clean sheets (100% save%) look pretty steady over the course of 120+ games but the number of <50% save% games that Hart has had increased from 2012/13. That is evident in Hart's rolling save% decline from 76.98% at the start of last season to 74.58% as of today.

It's interesting to look at the far right of that graph and see Hart's most recent games. Hart has posted a save% lower than his career average for the last 5 games. In those games Hart faced, or watched, 17 shots on target and he conceded 8 goals for a 52.94% save%. Now, if you just look at the red line in the chart above we know that such a low save% is out of keeping with Hart's City career.

This 5 game stretch cost Man City points, it cost Hart his place in the side and it enabled the media to suddenly write a whole load of shit regarding arrogance and character issues that we just didn't hear about when Hart was in top form.

Now, Hart's recent 5 game stretch, whether it be his mistakes, by eye or the numbers I am looking at, is admittedly poor. But how rare is such a five game streak in Joe Hart's career?

Hart's Rolling 5 Game PL Save%


Hart's most recent stretch of games is highlighted in yellow. Five game stretches where Hart posted a save% below 52% are highlighted in red. So bad streaks of save% have happened before and, save for a terrible run in the middle of last season (bins 87-101), Hart's save% tended to bounce back pretty quickly.

What I find really interesting is Hart's slow start to the 2012/13 season. In the first nine games Hart conceded 10 goals on 29 shots on target for a save% of 67% (the first five games of the season saw Hart concede 8 goals on 14 shots). It was a slow start to the season, City were conceding goals, Hart, by memory, didn't look great (beaten by the same shot type twice at home to Southampton),

But after that slow start, where it was evident that City don't give up that many chances, or chances of high quality, Hart's save% positively regressed, as expected ( 1 goal conceded in the next 5 games).

The point I am working at is: Hart has had bad runs before and bad runs at the start of the season and recovered in the following games. His save% simply regressed. This could be something that goalies go through, some bad games followed by a return to 'normal' form. The reasons for such swings: variance, poor concentration, confidence issues.

The problem with assuming that Hart will just bounce back is that he has a pretty poor save% record in the last year or so, caused mainly by that sub-par run in the middle of 2012/13. Are parts of the 2012/13 evidence of terminal decline or are they evidence of an extended poor run of form? Is some of the poor save% caused by the quality of chances City have given up - chasing games, no Komapny etc?

I don't know the answer, but I'd bet, with Hart being 26 years old, that it's probably not terminal decline. Hart has a career save% at City of ~75%, but since the start of 2012/13 that save% is 69%, or a fraction below league average.

Man City need better than 69% and Hart has proven he can be better than that number but during this season it seemed to me that he was fighting with himself and his game. He seemed to be over-committing in an attempt to impress and prove to himself, and others, that he could help his team in a visible way.

But in the process of trying to do too much he ended up doing too little - his job. Hart needs to stop the shots he is supposed to, come out for the through balls he is supposed to and leave everything outside of those bounds to the defenders in front of him. Instead Hart wandered for crosses (Cardiff), over played free-kicks (Villa) and rushed out for balls when defenders were in control of the situation. These rash incidents only further damaged the players confidence.

When Joe Hart returns ( v Tottenham) he needs to simplify his game and not wander out of the bounds of his comfort zone. Don't try and do everything, trust your defenders (a settled back four would help) and do not overplay any situation, whether that is a through ball, a cross or a shot from distance.

As a former goalkeeper, I can really sympathize with Joe's predicament. But he has to operate within his comfort zone and excel at it. Any efforts to do too much outside of his comfort zone will likely lead to mistakes and those mistakes are further driving down Hart's confidence.

Joe Hart's save% has hovered around 75% for over 3 seasons (it was ~79% during his loan spell at Birmingham) and it may be a pretty good guess that 75% is his "level". Given time, some work on his confidence and his ill-advised tendency to try and make an outstanding impact on the game, I think Hart will return to a level consistent with his career save% to date.

Yeah, I know.