paddy kennyAnyone who has been impressed by Liverpool’s progress this season and is wondering how they’ve managed to turn themselves into genuine title contenders after so many years as a Premier League also-ran should check out the film Moneyball.

The film is based on the story of the Oakland Athletics, an American baseball team in a similar position to that of most Premier League clubs insofar as, they can’t compete with the salaries and transfer fees the likes of Manchester City and Chelsea can throw at superstar footballers.

Instead, the Oakland Athletic’s adopted a purely statistical approach to player recruitment, forgoing the conventional wisdom of selecting players based on the subjective opinion of coaching staff and experts.

The Oakland A’s purely statistical approach took all available data, ran it through some complex algorithms created by mathematicians and recruited less glamorous players who the numbers told them could compete against the big boys.

Did it work? Well… yes. Hence the film. The Oakland A’s qualified for the play-offs with their collection of less glamorous, statistic-backed players with a wage bill that was only 1/3 of baseball powerhouse the New York Yankees.

This is exactly what the owners of Liverpool have attempted to replicate in English football. They have the advantage of being a big club who can afford high salaries already, but their recruitment policy now centres on the power of statistical evidence instead of subjective opinions, aiming to recruit undervalued players who can compete at the highest level.

I’m sure every football fan can name at least half a dozen players they feel are/were hugely underrated. Without the hype of the press and football experts comparing them to Lionel Messi every time you turn on the TV, their costs tend to remain a lot lower. Football is a game that rewards hype and effective self-promotion, no one demonstrated that better than David Beckham. Once someone has established themselves as a household name, their value tends to start increasing overnight.

So as Massimo Cellino prepares to takeover at Elland Road and revitalise an underfunded Thorp Arch academy, he too could benefit greatly from a more statistical driven approach.

With attention slowly turning back towards football for Leeds United, The Scratching Shed will be publishing a series of articles over the coming weeks which aim to analyse the performance of our existing squad, starting with the goalkeeper Paddy Kenny.

The reason I chose to start a statistical breakdown of team performance with Paddy Kenny is because he’s someone who I’ve often been disappointed with this season and I wanted to see if the statistics backed up my opinion that Leeds United would benefit from a new goalkeeper. Another reason I chose Kenny is because he’s been around a few years and would be considered a household name to many, he’s also one of the club’s highest earners and at a time when Leeds United are losing vast amounts of cash every month, the Moneyball approach really makes you think. 

The following infographic compares Kenny’s performance this season with the teams below and above us (ie. Huddersfield and Reading) in the Championship who share a similar defensive record to our own. The statistics I’ve looked at are the average amounts of saves made per game, the average amount of saves per goal conceded, distribution success (percentage of successfully completed passes, goalkicks and throws) and claim success.

So what do you think? Do Leeds United need a new goalkeeper? Is the Moneyball approach something we could benefit from? Take a look at the stats and let us know your thoughts at the bottom. 

Goalkeeper comparison

  • Michael

    Yes. I think the stats are pretty conclusive. Smithies is one of the worst goalkeepers I’ve ever seen.

  • Anthony Dickinson

    Stats can work if used in a purely unbiased fashion. For the Kenny analysis, we also have to take into account the defence in front of him, verus the defence in front of Smithies and McCarthy, possession and shots from the attacking teams etc. What’s the old saying, stats, lies and dammed stats!

    • TSS

      The stats do take that into account, the three teams I used have a very similar defensive record this season, but Kenny is saving less shots per goal conceded than the other two.

      He’s only making 1.93 saves per game (also less than the other two), the defence are letting far less chances reach him than Reading’s and Huddersfield’s.

      • Anthony Dickinson

        If you have the time, would be more worthwhile to see it across the whole of the Championship..

        • TSS

          I think the teams with the closest defensive record to our own above and below us makes a perfect comparison point, but you’re welcome to look up the other stats yourself if you think it’ll make an alternative argument.

          I’d suggest the most important stats are the two at the top, though a goalkeeper with better distribution is always helpful too since this relieves pressure on our defence.

  • badras

    this only works if they have had the same amount of shots against them . if 1 only had 6 shots and the other had 25 against them the chance that the 1 with higher amount could save more . then theres how good were the shots were they on target

    • TSS

      It’s taking the entire season to date into account, so the amount of shots they’ve had against them is a lot. Then it comes down to how many of those shots they save per goal conceded. The sample size is huge.

      Rather surprisingly, Kenny has actually had less shots to save than the other two this season, yet conceded a similar amount of goals.

  • jack

    So did Liverpool use this wonderful system when they forked out 50mill for Carroll?

    • TSS

      Not entirely sure to be honest with you, I know they’ve been bringing it in more with every season but I’d suggest Rodgers is key to it’s full implementation, they needed a manager willing to go along with the idea.

    • LUFCforever

      Pretty sure Carroll cost £35 million

      • Snowjoke

        That’s more than our whole Club!.

  • Danny

    I always thought Kenny might have poor stats simply because of his height. He saves everything around his body really well and for that reason looks solid, but when anything goes near the corner he gets no where near it, whereas taller keepers like McCarthy often claw those shots out, but when Kenny doesn’t get close to those same shots, people say he couldn’t have done anything about it!

  • yorxman

    Yeovil do…cheers fella!

  • White Rose

    Never should have sold Schmeichel. Horrible having to think about all the great players sold in the Bates era, if we hadn’t sold them we could have had a team of Premier League players by now. We must have sold him for under half what he’s worth now.

  • LeedsHipPriest

    Without a shadow of a doubt, YES. Would love to see McCarthy back at ER, but that’s not going to happen any day soon.

  • kennyshouldgo

    i dont understand why so many fans arse worship kenny, hes the most overrated keeper I can remember us having

  • Thommohawk

    I still think we badly need a new defence before we need a new goalkeeper. A mixture of athleticism, pace and physical presence! Hopefully as well a leader at the back with a football brain to marshal the back line. The current LUFC defence possesses none of these characteristics.

    The problem with Paddy Kenny is that while he’s had his off days as well there’s too often that he’s completely over exposed to opportunities for the opposition to score – the Sheffield Wednesday game would be a perfect example where 1 defender plays 6 players onside in the penalty area. No keeper in world football could stop a goal from those situations.

    So I think we need money investing in the defenders at this point. We have a good goalkeeper, brilliant strikers, average midfield and a relegation fodder defence I firmly believe that.

  • Sammy

    I think we do need a goalkeeper, Kenny seems to be letting in a few soft goals this season, and he’s also aging. In saying that he has had a few good games, but with him being one of the top earners at ER, we can afford to bring in a younger, and much more athletic keeper for the same amount he’s on currently.

  • Matthew

    Yes but only from a common sense point of view, while I’m happy with Kenny at present, he isn’t getting any younger. If we don’t think ahead now, we could end up with some random keeper who isn’t much good.

  • john palmer

    ALLthis latest stat crap its whats wrong with british football a headless chicken running in all the wrong places &passingthe ball 1yd backwards looks better on paper than aquality player using his brain eg Austin in team Diouf not!

    • TSS

      That’s a very extreme example, Diouf and Austin shouldn’t really be compared like for like, one is an attacking playmaker, the other more of a defensive midfielder. it’s like comparing Messi with Carrick. The stats aren’t the problem, it’s how some people choose to use them.

  • Perryman

    Yes he’s crap has cost us a few points qpr, Watford, Ipswich to name three,can’t believe mc Dermott keeps picking him,we wood be in the play offs now.

    • Matthew

      He’s won us more points than he’s lost, that is a fact.

      • TSS

        When you say “won us more points” do you mean, he did his job well some weeks? Because it is his job to prevent shots hitting the back of the net, is it not? That’s not winning us points, it’s doing what we pay him to. The stats show he has less saves to make per game than McCarthy and Smithies, so the defence are doing their part.

        I don’t think anyone is disputing he has some good games, but we demand consistency of every other player on the pitch and if they have a bad game they’re slated. Never happens to the keeper.

        Goalkeepers get a very easy ride in my opinion, they make a good save here and there and are praised to high heaven, yet when they don’t make the saves fans blame the defence.

        Penalties are the same too. Only the striker can possibly get criticised, the goalkeeper can just stand still and no one would blame him.

        So it comes down to not whether he’s done his job well some weeks, but how often he isn’t doing his job well and I agree with Perry that there’s been far too many occasions this season where I’ve felt Kenny could and should have done better,

        But that’s just my subjective opinion. I feel the stats say a lot more personally.

        • Matthew

          Kenny is a decent enough Championship level keeper who has played with an inconsistant defense infront of him all season, he has performed saves to keep us in games we really could of lost had it not been for his efforts, with his defense failing some games he has stepped up on more than one occasion. He’s not a perfect keeper but he’s a damn sight better than most keepers in this division and easily isn’t a player that deserves calling out for a few bad performances, we have a lot worse players in this team that seemingly get picked every week.

          Championship level keepers also aren’t a pinnicle of consistency, no keeper in this division is perfect, every keeper has had a few bad games this season. Stats don’t tell the whole story, they tell a story sure I’ll give you that.
          I don’t agree with some people getting the pitchforks out and singling out players for their mistakes, it’d be easy to do it for quite a few players in this side and stats be damned, most of these players wouldn’t make it in the premier league, nor would they make it in a Leeds side that was properly built. For the money Kenny(Besides say Schmeichel) is the best you’ll get with the money owners of this club were willing to pay in the past.

          • Matthew

            Also, while I’m one of few people here who will genuinely defend his performances, as far as I’m concerned he is money well spent and has put in a shift for the club and should be respected for that. He easily has another season left in him but should he be released at the end of the season or moved on I would respect that. There are easily more than a half dozen players I would rather see leave before him.

          • TSS

            I don’t know how you can point fingers at defence when he’s having to save less shots per game than Smithies and McCarthy.

            As for money well spent, I dunno personally. He’s one of the club’s highest earners, if not the highest earner (thanks for that Neil) and he commands such figures because of his name and the hype that comes along with him.

            You’ve said yourself Schmeichel was better and he was earning less at Elland Road. I’d say there’s a fair few other keepers in this division who are better than Kenny and earning much less too, now that probably wouldn’t have been the case a few years back, but he’s well past his best IMO. He’s one of those ‘jobs for the boys’ players Warnock brought in that severely jacked up our wagebill.

            The thing with some players (at least this is how it seems to me) is that they’re characters who fans take a bit of a shine to, ignoring their faults and hyping them up beyond their actual ability (Prutton and Hughes were also likeable but a bit meh in terms of ability). Kenny seems to fall into that category, always blameless despite being beaten at his near post so many times this season I’ve lost count, terrible distribution and the disadvantage he has when it comes to crosses putting pressure on the defence which a taller more athletic keeper would better deal with – (it may interest you to know that while Kenny beats McCarthy when it comes to claim success, he attempts considerably less than McCarthy does).

            But anyway, everyone can have their own opinions. I tend to take my own subjective view of a player and see if the stats back it up, which in Kenny’s case, they most certainly do. Smithies is a terrible keeper who is somehow making more saves per goal than Kenny is, I think that says a lot personally.

            PS. We have to get out of this Bates-instilled mindset that “average” is somehow OK for Leeds United, we should have the best of every position in this division. Players who can make the step-up to the Prem, not players who’ll “do a job” for now. That should never be good enough at this club.

  • Irving08

    Blimey Reading’s defence must be poor – that’s a lot saves their ‘goalie is having to make ! Just shows what a good defence we’ve got.

    • TSS

      Yeah, the defence was performing well for first half of this season, up ’til the recent bad run of form. Rare anyone praises a defence though because no matter how good they are, they all get beat sometimes so fans have the benefit of hindsight moments we can use to hit them with.

      Think the key to Leeds’ earlier defensive form was Zaliukas. He hit a bad patch and everyone else seemed to follow suit.

      • Irving08

        I think the main reason for the bad run was the fact that opposition Managers worked out how to defeat Brian’s 3-5-2 formation (as it was always likely they would.)

  • johnk

    He is by far the weakest link for Leeds, his distribution is rubbish and he never knows whether to come for crosses or stay on his line!!! Cost us more points than he does earning us points!!!

  • Stoney67

    Surely the point of moneyball is the financial cost? We all know McCarthy is a good keeper but would he be cheaper than Kenny. His transfer fee would be higher because he is younger (and better) and probably his wages would be as well.

    • TSS

      I wasn’t suggesting we sign McCarthy, but I think the point of Moneyball depends on the club. For Oakland A’s it was definitely the cost in cold hard cash, but to Liverpool I’d argue value for money is key since they can pay pretty high wages anyway. But why pay £50m for a midfielder when you can find one for £10m who’ll earn half the salary and perform just as well, if not better?

      I just think it’s a common-sense approach to transfers personally (so it’ll never take off in the Prem)

  • Joedqf

    The problem with this ‘moneyball’ approach is the number of variables in football, which is MUCH MUCH higher than that in baseball. Saves per goal, for example, doesn’t give a very full picture of a player’s quality at all. Let’s say that a team likes to consistently defend with a deep line, in order to stifle service to opposition strikers. Let’s say this forces the opposition to shoot from further away, where they have less chance of scoring and therefore a keeper gets to make more saves per goal. This is one permutation. There are infinitely more out there.

    In any case, Liverpool’s statistics based approach gave them such success stories Charlie Adam, Sebastian Coates, Andy Carroll and of course, Stewart Downing. Let’s think about that before we start praising it as the be all and end all of player recruitment.

    • TSS

      Liverpool are 4th in the Premier League and destroyed the league leaders yesterday. I think the change of manager has put a bigger emphasis on statistical player recruitment, Rodgers is perhaps more open to it than previous Liverpool bosses.

      As for less variables in baseball. Not really. The way teams pitch creates as many variables as you’re going to get from strikers shooting and creating chances, there’s variables in every game, but you play enough games and it creates a high enough sample size that the average gives a better indication. Take for example the high line you speak of, sure that may keep some teams pinned back but then you’ll come up against another with a pacey striker who’ll have a field day. Same striker wouldn’t have so much fun playing against a deeper line that he can’t easily gain a yard on.

      Point is, play enough games every keeper is going to face shots from all angles and distances, so the average shots per goals saved holds strong in my opinion. But even if we were to take McCarthy out of the equation, Alex Smithies of Huddersfield is still saving more shots per goal than Kenny is – how do you explain that one? Huddersfield aren’t anything special, nor is Smithies.

      • Joedqf

        Liverpool are obviously a great team at the moment, but they have probably the highest spend for the worst turnover of players of any of the big clubs. In addition to the ones I mentioned above, you could have Robbie Keane, Christian Poulsen, Alberto Aquilani, Paul Konchesky, Fabio Borini, Nuri Sahin, Assaidi, Iago Aspas. Their list of expensive flops in the last 4/5 years is uniquely terrible. All of which make me think that the good signings (basically just Suarez, Sturridge, Coutinho) have come out of throwing loads and loads of money at “maybes”, rather than brilliant transfer nous.

        Having said that, I’d be very surprised if the likes of Arsenal, Chelsea, Man City and Scum haven’t also been using stats very effectively for years.

        On Kenny, I wouldn’t say he’s an amazing goalkeeper and your stats are probably on the money. Was more trying to make a general point, but fair play in terms of your response.

  • PAUL W

    Yes, Leeds do need a new goalkeeper and it’s just a shame that Ashdown has been injured and Cairns is too inexperienced.
    Kenny has had it easy this season, as far as goalkeeping competition is concerned and he is often rooted to the spot when some of the goals have been conceded, with him not even bothering to dive to stop the ball from going into the net.
    Kenny may have made some good saves in the last two seasons, but he has also made many howlers, that have cost Leeds valuable wins.

    • Snowjoke

      Can’t argue with all the posters who’ve pointed out that Kenny’s a great shot-stopper. But that’s only one of the qualities we look for in a Keeper. Because of his physique, Kenny lacks several others that are arguably more essential. For example, the most solid defences always seem go have a big, dominant figure beteen the sticks – someone who not only bosses the goalmouth, through physical presence and force of personality, but has enough reach to deal with the fashionable “curler” heading for the top corner of the net. By reason of his limited stature, PK scores poorly on all counts and, imho, his talent for keeping out the short-range stuff isn’t really much compensation. Sorry, Paddy. It ain’t personal.

  • maxwatson

    I agree with the conclusions but not how you got there; I’m sceptical “moneyball” works in football, where play is extremely fluid and many factors are unmeasurable. Liverpool’s owner is John W Henry, the Red Sox owner, who appears in the film trying to hire the Brad Pitt character; the Red Sox went on to use the same approach and finally won the World Series in 2004, after an 86 year gap. They’ve won it twice subsequently.

    But the Henry era at Liverpool has encompassed the acquisitions of, among others, Andy Carroll and Stuart Downing at very high prices. The “stats” approach was meant to take a prolific scorer (Carroll at Newcastle), who – statistically – thrived on crosses and put him with statistically high crossers of the ball (Downing and Adam) and another statistically high goal creator, Jordan Henderson. None of them worked out at the time, three have left the club and Henderson is still there, doing statistically OK (6 assists and 1 goal in the premiership this season) but not lighting up the sky.

    Anyhow, Kenny. Difficult to gauge from the stats, because the defence has often appeared woeful. I’d say he’s a great shot-stopper but his distribution is terrible (maybe reflecting a Neil Warnock aspiration to fire it upfield). The thing that seems to be the issue generally with the defence is not the quality of players (I think all 4 central defenders are decent, Byram is maybe a better attacker than defender but not terrible, Peltier is decent for the Championship and we have arguably 5 passable left backs if one includes Taylor) but the organisation and lack of leadership. I assume that is what Zaliukas was brought in to do, and why he was given the central role in a 3 man defence, but he was obviously terrible at it.

    In the O’Leary team, Radebe and Martyn both organised the defence. Radebe raised the game of any player playing alongside him. Then Radebe went on the long term injured list and Robinson forced his way into the team – and I’ll bet a lot of his statistics were better than Martyn’s. Martyn left for Everton, who perennially hung around the relegation zone at the time. Everton stayed up. Leeds went down.

    Alex Smithies, by the way, looked very vulnerable from crosses and set pieces in the Leeds game, a fact belied by the stats.

    Anyway, rambling aside, I think Leeds could do with a goalie who operates as part of a well organised defence, and I agree Kenny isn’t it. Have a look for someone in their 30s who has kept goal for Italy. Maybe Buffon (aged 36 now) is up for a move?

    • PMH

      I agree with your last point. Defence is a team thing, and Leeds defence looks panicky under pressure. An experienced and sound goalkeeper-central defender pairing is crucial. Fabulous goalie saves are nice but not really the key to success.

  • Tim Baker

    Stats should certainly be used more in football. This is the world’s richest sport yet remains a game of opinion rather than fact. I agree with a lot of comments about the “fluidity” of football. It is safe to say that most American Sports are a series of set pieces where the field is set and the play begins and ends (see American Football and Baseball). It could be argued that basketball is the most similar to football but with play changing so frequently and the field “set” the same argument could apply. It is therefore a useful exercise to analyse the statistics for set pieces but less relevant for open play. This would be better than the current analysis.

    • TSS

      Yeah, I agree the fluidity thing is a good point. You can measure how well someone passes and shoots, but there’s an x factor that’s much harder to determine in how players work with one another. I’d say the use of stats in football is still rudimentary and there’s more and more measurements they can add beyond the obvious for things like distance of a players pass, how he moves, where he’s passing the ball (always backwards in the case of Kilkenny for example) etc that’ll help narrow down how well he’s going to fit into a team, but there’s going to be a necessary human element for some years to come.

  • Ev

    Ive read peoples comments and I think as usual the truth is somewhere in the middle. I think yes Kenny is not playing particularly well at the moment, hes probably lost a bit of confidence but hes also been questionably our MOM on a few occasions so for me I would only replace him if we had lots of money to spend. McCarthy is a fantastic goal keeper. He was Readings best player when they were in the prem and for me hes one of the hidden gems along with Tom Ince and Will Hughes that Prem clubs will eventually cotton onto. The defenders are a problem. Both pearce and lees are steady markers and good in the air but its their distribution that’s the key issue. Im sick of watching Lees just hoof the ball up (especially if Smith is playing). He obviously got adapt his style of play so central defence might be an area where we could do with a few Italians mr Cellino.
    Shopping List
    A dominant midfield general, 2 centre backs and a striker

  • PMH

    “Moneyball” applies to baseball which is an individual sport with a lot of recorded stats. It won’t work so well for football. Nevertheless, players are a business commodity and the usual business rules apply. One way of thinking about professional football is as a means for developing player value. You buy prospects cheap, build them up, feature them in your team, and then sell them on at inflated prices to teams with more money than sense. It certainly beats the alternative plan which weak managements get pushed into by fans: lose a few games; pay over the odds for a big name player; when the player has a bad spell; blame him and sell him on at a discount. (This is what Leeds seem to be doing with Diouf, by the way.) The problem for Leeds is going to be that Premiership teams will be poaching their best players continually until they can get out of the Championship. Selling those players occasionally, at a good price, makes business sense, but it won’t help the team get back into the big time.

  • john palmer

    reply to tss Austin & D iouf are areal life example ,both play behind a striker & my point is if picked on area covered shots passes tackles AUSTIN would look better than DIOUF but if you know anything about football you can see Austin is not fit to lace his boots. with Messi if anyone had to look at stats to see if he is better than another ,god help em

    • TSS

      Diouf is a wide player/AM/second striker, I’d compare him only to someone similar. Not to Rodolph Austin who plays in nothing close to the role Diouf does.

      Austin is a DM we’ve been playing further up the pitch recently, so you could maybe describe him as a box-to-box midfielder, Diouf on the other hand is more of a number 10.

      Quite what this has to do with comparing goalkeepers (who I’m sure we can agree, definitely play in the same position?) I no longer remember…

  • Colin

    Harsh on Kenny I think. I reckon if you asked the other 23 teams in the Championship if they would rather have Kenny than their own keeper, I bet more that 50% would say yes.

    And as for stats, I hope we don’t count on them too much. Ross McCormack had a goals per game ratio of 0.15 last season, so on that basis we should definitely have got rid of him. Now his ratio is 0.73 goals per game.

    We’ve had some ropey keepers in the past. Personally, I’m quite happy with Paddy Kenny.

  • Nick Ryle

    Kenny was poor last night. Slow to come out a couple of times, below par distribution and looked a bit overweight. I think he’s been OK up to now, but definitely time to change in the summer. Anyone got a real feel for how good Ashdown is? PK could certainly do with some competition.

  • lufcboy

    Yes – Kenny’s a bit of a liability.
    Was it not rumoured that Cellino would bring in Julio Cesar?

  • paul

    Paddy Kenny is far to slow off his line and is rooted to it on every cross. When teams cross a ball into the box he should be out to meet it not rooted on his line hoping the defenders can deal with it. The Leicester game was a good example the ball that came to nuggent was on the edge of the six yard box and he should have easily got to it and made it a harder opportunity, had that been Schmeichel at the other end it would not have been the same result. Kenny has had his day, we need a keeper in his twenties whos on the way up not down .

  • Pingback: Shock as Whites sign player of genuine quality | The Scratching Shed()