While doing research for elite goalscorer stats over the past few seasons, I finally had a chance to really dig into the numbers and do a full set of conversions to control for penalties and time played.

Why would you want to do this? Two reasons, really…

1)      Penalties are converted at a 76% clip on average anyway, and any number of players on a team can do this. Therefore you shouldn’t be rewarding guys for scoring penalty goals – you actually need to subtract those from goalscoring totals. Thus we start with non-penalty goals.

2)      Not all players are on the pitch the same amount of time. Some get subbed off regularly, some are only substitutes, some get injured, suspended, etc. By doing some simple calculations, we end up with non-penalty goals per 90 minutes played or NPG/90, which is how all goalscorers should really be evaluated, both in the past and going forward.

Those drawn here by the headline are all thinking, “Shut up, stat man and just get to the point.”

Well, once you do the conversions, you discover that despite the monster goal tallies, Lionel Messi and his superhuman footballing prowess has actually only been the most impressive goalscorer in Europe in two of the last four seasons.

Additionally, you might figure that Cristiano Ronaldo has consistently been the second most impressive goal scorer in each of the last four seasons…

Well, that ain’t quite true either.





Messi has lead the world in NPG/90 twice in the last four years. He also posted a second and a third, but it wasn’t Ronaldo who clipped him to the top spot in 11-12, it was Gonzalo Higuain. Meanwhile, Ronaldo was actually tied for 1st in 2011 (with the Gomez Button), and then second in 2013, but 3rd and 5th the other two seasons in question.

This isn’t to say that both Messi and Ronaldo aren’t amazing – they clearly are (and Messi’s value will become even more apparent when I talk about G+A/90 in the near future). They are also the only two who have been in the Top 5 the last four seasons (Higuain was 12th one of those years). It’s more to highlight the point that there are other players in Europe who have also had amazing goalscoring production when controlled for certain important factors.

Just looking at the headline stats for football can be misleading and costly, both in terms of paying too much for transfers, and when teams undervalue an incredible scorer like Higuain and let him slip away. (Here’s looking at you, Arsenal.)

Anyway, I’ll be going over a much broader list of goalscorers for each of the past four seasons in the next week, as well as highlighting different ways of evaluating scorers, providers, young players, and transfer targets.

If you like this, check back soon – there’s a lot more to come.



  • toshack


    Following you for a while it’s good to see your long lasting praise for Higuain come good – in stat terms anyway…
    But if memory serves me, well, well or thereabouts I seem to remember you had Lewandowski very high up your charts last summer? And here he’s only made the 12/13 chart. He didn’t go back further than that? Or is my memory as mushy as the peas I get with the fish ‘n chips when visiting the UK?

    Btw your follow up on last summers hot transfers that you promised – did I miss that or is it still to come?

    • tknutso

      This stuff is the precursor to the follow-up – a lot of work to establish baselines for performances that we should be interested in and put them in a format that strips out some extraneous values.

      Lewa’s stats and that write-up is here.

      I _just_ cut to top 5 today. Will be much more detailed with baseline stats across leagues and previous years in the near future (mostly because I finally finished collecting the data and we don’t have to add more new info every time they play a match like in the current season).

      • Toshack

        Thanks Ted,
        Looking forward to what’s to come!
        Happy to see my memory was ok with regards to Lewa ha, ha…

  • http://blogs.columbian.com/portland-timbers/ Chris Gluck

    Super article Ted and great kudo’s for you on your research! Just curious have all those top players played for the same team in the last four years and perhaps the other filter (if time permits) is to arrange that data by League?

    I remember we’ve talked about how the team impacts individual player performance and how the different leagues have different skill sets/playing styles.

    It would be interesting how those filters shake out the top 5 strikers for each league… anyhow; again GREAT stuff Ted and not boring from my viewpoint….

  • Dave

    I like the metric. Although the amount of playing time should matter as well (perhaps with an accompanying metric, or perhaps just minutes played). For most, performance decreases with additional playing time (due to limited recovery times), and some players are more prone to injuries, etc. Thus, one should not pay a “Messi” salary if the Messi-like performance level will not continue over the full season.

    But I would rather start with an apples to apples comparison such as the NPG90, and then look into the crevices.

    • tknutso

      Dave: amount of playing time is right there in the graphics.

  • http://www.wengerball.com wengerball

    Although removing penalties & working over 90 minutes is better than using all goals, sometimes I really do wish you could factor fatigue into those numbers.

    I would argue that coming on as a sub in the last 25 minutes gives a forward a huge advantage over defenders, both mentally & physically. Consequently would expect a subs per 90 numbers to be marginally better than if he’d been on for the whole 90 minutes.

    Then there something to say about game state.

    Anyway that just a roundabout way of saying I’m still not convinced that Higuain is as good as the numbers say. He’s good but not *that* good.

    • toshack

      It’ becoming a bit difficult, cause now you end up in a discussion of defining what “good” is? Higuain sure didn’t score the kind of “great” goals that Messi or Ronaldo did/do, but when Higuain was playing on the pitch for Real he sure scored a lot of goals. What more can one wish for?

  • danny

    nice job.

    i think the next thing we should take into account is the game state. some goals are just more important than others and easier (sic) to score. e.g. 3rd, 4th, 5th, 6th goals in a 6-0 home win etc.

    • tknutso

      I disagree with a lot of the concept of “goal importance” when it comes to individual assignation, so that type of analysis at the player level would be unlikely to come from me.

  • alexander.radtke@gmx.net

    Hey, great work there!
    One idea popped up when I read your NPG90 index. I think saying “Therefore you shouldn’t be rewarding guys for scoring penalty goals” is a little bit too easy. There may be a reason some player shoots a lot of penalties because he may just have the best conversion rate (e.g. when you look at the penalties of Salihovic from Hoffenheim you can actually see he is better than most others in penalties). Additional, as we talk mostly about strikers there is quite a chance they would have made a goal without the foul that led to the penatly anyway.
    So what I would like to suggest is instead of counting the absolute goals looking at the conversion rate. So if a player attempted 10 penalties and scored 9 of them, he would have (0.9-0.76)*10=1.4 goals from penalties added. On the other hand they could have a malus, if their conversion rate is lower that average. Of course that results in fractioned numbers but I think it’s still a good way to introduce the players quality at penalties when we talk about individual statistics.
    What do you think about it?

  • http://blogs.columbian.com/portland-timbers/ Chris Gluck

    By the way, I fully support the logic of separating out goals from PK’s, indeed I wonder if there are three different categories to consider but will leave it to you given this research is down your alley…

    Run of Play Goals
    Penalty Kick Goals
    Free Kick Goals

    With the game having three distinct ways to score perhaps there are strengths in some players that shine better in re-start/set-piece scenarios than normal run of play conditions. While I’m not good on memory I’d venture to offer that Beckham was leveraged as he aged a bit more for his wicked set-piece free kick than his normal run of play ability – others may view that differently but however viewed Ted I will continue to watch how you progress.

    • tknutso

      Chris: I think that makes a lot of sense at some point. It might be that you slice stuff too thin eventually – how many free kicks are really taken in a season by one player, etc – but thus far we don’t know. Just takes time to spit out and explain all of the data at the moment.

  • Pingback: How to scout goal scoring talent? | 11tegen11()

  • http://nextfifapresident.blogspot.com Steve

    Interesting analysis. If we’re going to exclude penalties because they are considered gimmes, should we not also subtract one goal for any penalties missed? Just a thought. Don’t know how this would affect your stats.

  • Pingback: StatsBomb | From Awesome to Average to Awful: Goalscoring Stat Distributions()

  • Pingback: Alexander Lacazette and the New Generation of Olympique Lyon | Think Football()

  • Pingback: Attacking stats of AdriCiYang and Lewandowski – a first snapshot, part I | reepratio()

  • Dragão

    Firstly, i would like to congratulate you guys for the amazing job that you have been doing in this website.
    With that being said, i think that we can´t talk about the best goalscorer if we only analyze the national leagues. National cups, international games with national teams, we can neglect all those competitions. But leaving UCL out? The toughest and most important competition of the game.. It’s like electing the best swimmer, runner, cyclist without taking into account the gold medals at the Olympics..

Improve Performance and Productivity in Your Club:
State-of-the-art Football Analytics