I did a bit of research, and this was one of the better opponents he played that year (difficult to compare with club teams - World Soccer noted in 1964 that club teams already tended to be better integrated and attuned than national teams). It was certainly the best opponent of that year (and perhaps his entire national team career) where he scored 2 or more goals against. He did that in the other match in Moscow, rather than these 50 minutes highlights. I hope you understand my point of the post above, too (- there one can notice who are the real underdog footballers).
Yeah, I assume (although the apparent discovery of a left footed volley that went just wide in the Maracana game if I spotted it right on another video, puts it in slight doubt maybe before I try to check at all) that the skill moment on the edge of the box where he plays the ball through his own legs must also be in the away game (but I saw it on compilations rather than the highlights of that game IIRC...which were much less than 50 minutes worth). Yes, I understand your other point/concerns. No criticism of annoyedbyneedoflogin's ideas, but of course if the aim is to look at performances critically and pick out losses of possession etc (which don't even need to be mistakes in some cases) then most performances, including very good ones, will likely have quite a few. I saw quite a bit of Barcelona last night and despite the close score I did think the likes of Messi, Iniesta and also before he was taken off I'd say Coutinho too were playing skilfully and positively etc, but there were certainly moments they got dispossessed or a move broke down on them too for example. Anyway, I think annoyedbyneedoflogin wants to modify his system, but I suppose it was an attempt to statistically differenciate between some great performances (he said himself that Cruyff in the Arsenal game was in 'Di Stefano mode' in terms of involvement, drive etc, so that's a positive observation but not really what the system was accounting for unless it involved taking multiple players on - I didn't check and verify properly whether I thought there might be some moments that qualified on that basis or not myself anyway...but I guess there'd be a few or more in the other Cruyff performances that I myself listed anyway, even if going past just a single player or bypassing a single player with a pass can at times be very useful/crucial in building up to a goal and require great skill or vision too anyway.
I added weights to the formula. Formula = (Flaw+0.5Lack-2Bonus)/1.5Time So the lower the score, the better. Some examples: Maradona vs Belgium 4,1,1 over 4 minutes. 2.5/6=0.42 Di Stefano vs Frankfurt 4,5,1 over 6 4,5/9=0.5 Rai vs Barcelona 7,3,1 over 5 6.5/7.5=0.87 Cruijff vs Arsenal 15,5,0 over 13 17,5/19.5=0.90 Pelé vs USSR 13,4,3 over 6 minutes 9/9=1 Please note that the numbers aren't precise. I roughly estimated time as whole minutes. The Pelé video is cut really short, maybe it should get an adjustment. Also my observations show different results from @PDG1978. However, the main goal is to refine the rating system without making it too complex. Firstly, it is more important to establish a system that provides agreeable comparisons overall than to agree on exact numbers between specific examples. Feedback is welcome
Over the full 90 minutes that would mean Pelé ends up between 1.50 and 2.00 (roughly). Hard to believe it is a factor four worse than the top. I don't want to be harsh @annoyedbyneedoflogin I get the general idea.
Firstly, it is more important to establish a system that provides agreeable comparisons overall than to agree on exact numbers between specific examples. Secondly, when a film is cut short, it means that it may have been longer. Adding time means a better score. Thirdly, factors are unimportant in this stage of the process. One can imagine the sums being substracted from the number 10 for example.
Yes, that means a better score. But it also means, in all likelihood, more moments that you would regard as errors or 'lack of creativity', while the so called rare 'bonus moments' don't increase that much, or even not at all (since it is to be expected that those 1-3 moments are included in the 53 minutes). Therefore, the score tends to be worse. The good moments don't increase at the same rate (in your formula).
There can't be a formula that can summarize a single performance, at least not a simple formula, which is what you are trying to establish i guess. There are numerous things you can do right or wrong and as far as understood, you mostly look at things that footballers do on the ball. What about things they do off the ball? As great Johan Cruyff said: "When you play a match, it is statistically proven that players actually have the ball 3 minutes on average … So, the most important thing is: what do you do during those 87 minutes when you do not have the ball. That is what determines wether you’re a good player or not." Lack of action (not recognizing open spaces that you can run into) or overacting (you choose wrong off the ball movements) are mistakes as well (which you don't see in individual highlights) and in my opinon, they are even bigger mistakes than things like getting dispossesed or "lack of creativity" or making a single bad touch. For example, Messi, imo, was absolutely amazing in last el clasico in that sense. He is so damn inteligent. He recognized that he is being man marked and he completely exploited that fact. If you notice how he moved, you'd understand that most of the things that Barcelona created in that match, came from Messi distracting Real Madrid defense in some way. First goal is an ideal representation of how significant that is: Here is another example in el clasico, which highlights how important movements actually are in football: That's something that your formula don't take in consideration whatsoever, and i hope i demostrated you in those two videos that not taking them into account is foolish because they are just too important to be ignored. If that's not complicated enough, you still have to consider that not every player has a complete role on the pitch, so to speak. I mean, they all have similar tasks like making an inteligent type of pass forward when they have a chance, but in tactical sense, all of them obey what their manager says, which can completely change the way they play. That's something that has to be taken in consideration as well. So Jordi Alba getting caught out of position on counters might not be his mistakes, but rather managers mistake for forcing him to get forward so often. It's just an example, i am not saying he does that, but you get the point. Also what might happen in defending is that someone else fails to do his job so you are there trying to fix that, but end up not fixing it, but looking like a complete fool. An average viewer would think that the guy who tried to fix the mistake made the mistake, but what truely happened is that he did his job, but was left hanging by his teammate improvidence. Defending is very tricky and a team type of work so you really depend on your teammates. I could go on with all the angles you overlook with such a simple formula. And although you might think i nit pick and that those examples are just rare exceptions, well you would be wrong thinking that. Football is not simple. It looks simple if it's done right, but it's not.
I think it's right what you say in general, although of course for the best performances ever what is done with the ball will be a huge part of any performance deemed a contender, especially for attacking players - sometimes good off the ball play results in players getting the ball and that can be appreciated from some of the compilations but I suppose the best judgement will be formed looking at full games and then everyone will have slightly different conclusions I guess even if quite similar in a lot of cases, while stats systems will always give different scores based on criteria chosen and weighting etc. I think in the full version of this FA Cup Final, the commentary team go on to praise Dalglish's run at around 7:35, for the goal from the move starting at 7:29, and that's a good example of what you're pointing out I think, although of course the great thing about football is that various good options exist (not only in this instance that Whelan could actually choose the pass to Dalglish instead of Rush, but that Dalglish might have run through the centre for a pass too - whether he had in mind if he went wide it'd open up the pass to Rush more I don't know but from what I recall - seeing it recently not only at the time - the commentary team might have thought so). In any case the brain is obviously a great asset at times, and making positive active moves even without the ball can of course be highly beneficial yep.
I am not denying the importance of things done with the ball just saying that off the ball has to be looked at as well with no exception. In overal it's intriguing topic and I am definitely interesting in helping to find the right balance in evaluating players, so i will propose the way i think things should be judged, but in few days, probably monday.
I think youtube compilations are entertaining more than anything else, but on average there are so many flaws to said videos. One of the indispensable factors that goes unseen is the context of how the other players play, for example, Cristiano Ronaldo's performance at the Champions League Final 2008--the youtube video is impressive, but when you watch the complete game, you notice how great Wayne Rooney also was, that is, not quite as great as Ronaldo was, but definitely close enough that it puts a serious dent on Ronaldo's performance. Also to consider is Frank Lampard's performance, which was arguably exactly as great as Ronaldo's performance, or at the very least close enough to Ronaldo's performance. Of course there are many other significant factors that go unseen; work rate, positioning, general influence compared to the other players in the game, etc. Another classic example of how youtube compilations can be extremely misleading: Zinedine Zidane vs. Spain 2006 (World Cup Quarter Finals). When you watch the complete game, you might (if you have a keen enough eye) notice that a lot of players were in fact better than Zidane. All of Ribery, Makelele, Vieira, and Puyol, were clearly superior and more influential than Zidane was at any relevant point in the game. Moreover, Xavi Hernandez before he was taken out of the match (at the 72nd minute), was arguably better and more influential than Zidane. (In fact, I actually think that it was a considerable tactical mistake to have taken out Xavi. Coincidence or not, France scored the 2-1 goal 11 minutes after Xavi was subbed out.) And then... Zidane went on to score a great 3-1 goal in the final seconds of the game, nearly exactly when the 93rd minute came around, when Spain logically had their entire team going forwards trying to snatch a late equalizer goal. And so, Zidane made Puyol look bad when the truth is that Zidane could have simply retained the possession of the ball for a few more seconds, passed the ball backwards or sideways as was his signature, and the game would've ended, because Zidane's goal was scored that late in the game, that the game would've literally ended in less than one minute--it was 3 minutes of extra time, and Zidane scored his goal when the clock had nearly hit that 93 mark. A beautiful goal by Zidane, but as beautiful as it was, it was more or less equally as pointless. Moreover, I think that it can be objectively demonstrated that all of Ribery, Makelele, Vieira, and Puyol, had considerably more solid performances than Zidane, who was an unequivocal passenger for 92 minutes of the game.
Without a doubt, off the ball movement and positioning are important aspects in football. Some may dribble the whole team to score a goal, others just stand in the right place at the right time. If you have any idea on how to incorporate it in a formula for distinguishing top performances, please share your ideas. I have further contemplated the current formula. There is definitely something to be said about the "Kaka pass". It sets defenders at a disadvantage, leaving an opportunity to score 1v1 against the goalkeeper. I believe this pass to be more individualistic and difficult than executing a cross or a straight throughball. Therefore, I agree with @PDG1978 that this may deserve a minor bonus (*1). Other possible actions deserving either this minor bonus or the full monty are: -Penalty saves, excluding shootouts. -Interceptions in offensive area. This is currently considered as a factor 2 bonus, under recycling possession while being outnumbered. The reason to lower this bonus is because it may (partly) be because of a defensive error rather than keen awareness that the attacker gains possession. Feedback and opinions for leaders not minions
As I mentioned before, there are games where several attacking players have good games (i.e. open fluid games that are generally 'easier' for attacking players), and there are games where attacking players are generally subdued by the defensive players (i.e. difficult games where the attacking players have a hard time making a positive contribution). In the difficult games I think that Factor A should have a higher score than Factor A would have in an easier game. To be clear: the difference between a 'difficult game' and an 'easy game' is not defined by the quality of the opponent, but by how 'difficult' or 'easy' the game is for the top attacking players of both teams. With that in mind: How could you determine how difficult or easy the game is, on the basis of watching the on-the-ball actions of one single attacking player? That is perhaps the biggest flaw with internet video compilations. At the same time, with legendary players, one can relatively safely assume that their positioning is generally world class, in which case just assessing what they do with the ball can give you, not a perfect depiction, but still a fairly accurate depiction of the performance. Overall, I generally agree with your methodology and I think it offers some useful insight. (At the same time, I cannot stress enough how essential it is to know how the other world class players compare to the one player you're trying to measure.)
So here we go. This is gonna be a looong one i think. I really want to get this thing done right. First things first, i really like an idea of negative points and an idea of "lack of creativity", but giving negative points for every backwards pass is insane. They are not necessarly bad at all. Football would ridiculous to watch if every pass was attempted to be creative. So no, i completely disagree with such concept. I was thinking on how to minimize subjectivity in evaluation (or get rid of it if possible), while not getting too obsessed with statistics like key passes,. and the conclusion i came to is that you just can't get rid of subjectivity completely and that the most accurate way of evaluating players (and in my opinion, only way worth doing it) is going play by play taking every play in its individual context. By "its individual context" i mean completely abandon statistical logic like saying that every goal worths +1 point that every key pass worths +0.5 or anything of that kind. Not every goal or key pass is the same and such should be evaluated... so how am i going to interpret all of them? The very essence of approach i am going to take is something likes this: Player's every conscious attempt, whether off the ball or on the ball, to do something is recognized as a single play. (Lack of attempt (if not being a significant mistake) will later be included in bonus points) Every play gets one of five values ("-2" or "-1" or "0" or "+1" or "+2"), -2 being the worst, +2 being the best mark. Clarification of values Majority of plays will get a value 0. Those are minor plays that have no significant value on the match. Every play valued 0 gets one of three signs ("-" or "o" or "+") For example. Play that's valued as 0 with negative sign, "-", is perhapse a lost duel in the middle of the pitch (50-50 duel that's won in the middle of the pitch is 0 with sign "+"). That play doesn't affect player's major score, but it gets accumulated and counted at the end of the match. If player, at the end, has 10 lost duels and 15 won duels his accumulative score (net score) is +5 (in terms of duels alone, they are other things as well). The next rule is that every 5 of something (in net score) counts as 1 in a major score. So in this case, player having +5 net score (15 duels won - 10 lost) would get +1. That +1 is counted along side values mentioned in point #2. If his net score was -7, he would still get -1 in a major score. If his net score was -10, he would get -2 in a major score. (((Whether 1 point in major score should be added after every 5 of something or after 10 of something or any other number is open for the debate))) Value 0 with "o" sign is a neutral mark and it makes no difference on overal score. (Tho, it might make difference in bonus points). When player gets -2, -1, +1, +2 values? -2 means that player made a mistake that lead to a huge chance for oppositions (Example video 1 *examples are in the post bellow)) or that player missed a huge opportunity to score (Example video 2 and 3). -1 means that player made a mistake that lead to potentially a huge chance for opposition (Example video 4) or that player missed an opportunity to make an open forward pass that could lead to big chance or he missed a big chance in front of goal. The trickiest part here is to differentiate a huge mistake worth -2 and a big mistake worth -1. +1 means that player initiated, potentially, a big chance with either his pass or dribble. NOTE: not every key pass is +1 and not every dribble is +1. They are rewarded with points if and only if it initiated a big play. +1 is also rewarded to players who made a very good shot that would be scored if goalkeeper didn't make a big save (example video 5). Also, goalkeepers get +1 for great saves. Players get +1 for tackles that potentially saved a big chance (for example. If a defender makes a sliding tackle to stop a counter attack. NOTE: goal saving tackles are worth +2) +2 means that player created a big chance for goal either via single pass or multiple dribbles. Unstoppable long range shots are worth +2 or very technically demanding shots like the one by Zidane in 2002 ucl final or Mandzukic's in last final. Goal saving tackles are worth +2. So to summarize, when play results in potentially a big deal it gets rewarded with 1 point (+ or - depending on whether play is positive or not). -2 is given to mistakes that result in huge chance for opposition. Analogously, +2 is given when something extraordinary has been done that actually resulted in a huge chance for his teammate or goal (if the shot is the extraordinary thing). NOTE: I can't stress this enough, the value you get doesn't depend on the outcome. 0 is essentialy for neutral plays, but the count of whether it is positive neutral or negative neutral is accumulated at the end. Another way to look at value 0 is in the sense of expected play. If an attacker shots on the goal in a way that you would expect an average attacker to shot on the goal in that cirucmstances, he gets value 0 regardless of whether goal is scored or not. What i mean by that is that in cases like this (0:20): Player doesn't get rewarded with +1 for that shot despite scoring a goal, because shot itself was nothing special and it was goalkeeper's mistake (goalkeeper gets -1). So in this case play gets 0 sign(o). Here is an example of how it should be done: Kaka's evaluation: +1 first touch #Kaka gets rid of a defender with first touch and that move initiates potentially a big play +2 pass #Kaka's pass creates a big chance for Crespo. it would be rewarded +2 even if Crespo missed it. So in that sequence of play Kaka earns +3 points, which is around the most players would get for a single sequence. Another example: Messi's evaluation: +1 dribbles #Gets rid of multiple players with impressive close control. Opens up potentially a big play +2 pass #It creates a big chance So, like in Kaka's case, Messi gets 3 points. Villa would get only 1 point (for an inteligent run). Finishing was an expected type of shot so no points there. Bonus points I had in mind giving bonus points for several things. An engagement Points being given in the same fashion: -2 very disconnected from the game (i am not sure anyone gets this low) -1 disconnected from the game (some players sometimes, but rarely) 0 normal engagment (majority of players should be here) +1 engaged (Busquets under Valverde is a perfect example of someone who is very engaged in attack and defense as well. He would get +1 point at the end) +2 very engaged (maybe Kante at his very best, but this would be a rare case) Difficulty This is very important. I would give it in the same fashion as engagment. -2 for playing very bad performing team and so on up to +2 which can be something like playing in UCL final or playing vs very great defense (if player is an attacker) and so on. Still thinking this through. But majority of players again should be around 0, which means no bonus points nor negative points, I have some more things to say, but i feel tired writing all this so i will leave it for some other time. Tell me what you think. is it waste of time or not?
Example video 1: NOTE: Cech would get -2 regardless whether Ayew scored this chance or not. Example video 2 (4:22): This would be counted as -1 if Neymar wasn't there open on the far post. Since Neymar was open, it is a huge miss because a simple square pass would result in a goal. It's an example of not passing when you should pass. The chance to shot itself wasn't that big so Cavani would be rewarded with only -1. It's still a big miss. Example video 3: Really big misses are -2. Majority of misses should be -1 as Cavani from example above. Example video 4: Messi gets -1 for losing possesion in potentially dangerous area. It's important to highlight that this would be -1 for Messi regardless if counter attack ended up as a goal or not. Generally, rule of thumb is that every misplaced pass or lose of possesion that results in counter attack is rewarded with -1. Example video 5: In this case Messi is rewarded with +1 for an excellent attempt that would have been scored if Navas didn't pull off an amazing save. Navas gets +1 as well (maybe even +2 debateable). NOTE: If Messi did score that he would still be rewarded with only +1. A result of a shot doesn't change player's effort. This is one of the key things to understand.
Given that a rep for the endeavour and because it'd be interesting to see the results on this basis for various performances, and it might be a helpful tweek that annoyedbyneedoflogin likes (of course he has to confirm that himself!). I won't try any critique of the ideas as it's a lot to take in and analyse, and also I'm probably still of a mind that individuals watching the games and giving their own scores (their brain having taken in every action) might be as good a way to make scores as any still (people having different standards is obviously problematic though if used for comparison), but it looks to be a statistical method that feasibly gives reasonable results at first glance.
At first glance, it seems like a relatively subjective system that could apply for full matches. To specify; How do we know if something is big or small? Is this equal for every position? How does a 3 minute video compare to a 9 minute one? These are just some of the questions that come up when you start rating videos and comparing them. Id say, just try and rate some of the videos on PDG1978 's list and compare the results
Im not sure if I mentioned this before, but I consider one two passes as creative play, as the whole tends to be a forward move, despite a sideways or even somewhat backwards pass.
The more I think about it, the more of a 'skeptic' I become. I mean, when you watch the complete game, you might notice the fact that Riquelme vs. Arsenal 2006 (the second leg, at Villarreal's stadium), well he played essentially as an anchor midfielder more or less, arguably more participative overall (read: aggregating both on-the-ball and off-the-ball energy) than Villarreal's actual defensive midfielder, Marcos Senna. At any rate, Riquelme had a significant influence on the possession and most importantly on the fact that Arsenal was defending a 1-0 lead instead of trying to extend that lead. Not to mention the fact that Riquelme was involved one way or the other in at least 2-3 clear cut chances, out of the 4-5 clear cut chances that Villarreal failed to score in that game. On the other hand, on the basis of watching the youtube video of that performance, the average viewer will probably (and in most cases, readily) assume that Riquelme was doing basically nothing off-the-ball, and that his on-the-ball contributions had no direct effect nor any measurable pre-effect on the many clear-cut chances that Villarreal failed to capitalize on. Which is generally what happens with youtube video compilations: detractors will fill the blank spaces with assumptions that consistently correspond to their bias against the player, and fanboys will fill the blank spaces with assumptions that consistently correspond to their bias in favor of the player, but you rarely ever get an actually accurate depiction of the performance. Youtube video compilations are simply too flawed in too many of the fundamental areas. Ultimately, rating a performance, CORRECTLY rating a performance, is time-consuming hard work, it's not easy work, and not only do you not get paid a single cent for your effort, but most of the time you have to also defend yourself against unwarranted accusations by the fanboys who can't handle the accuracy nor the honesty of your criticism. In conclusion: doing a half-decent job of it should always require watching the complete game, in my opinion.
Yes, exactly. It's pretty subjective because i simply don't see any other way of truthfully evaluating performance. Statistics just don't do justice so all has to be taken into a context else it's waste of time. This is actually how i evaluate players in general when i watch football, except i have never put things in actual numbers (so far), but rather at the end i rate him based on gut i developed watch the game. This system is mostly about putting thoughts and feelings on the paper and materialize it in the sense of numbers. I am aware of the fact that others may read context differently in some plays, but perhapse that's a good thing. This is another way to look at all of that of what i was saying (i think that's pretty useful way of looking at performances): Always, in the back of your mind, keep track of what an average player would do in a certain play. If player that you are trying to evaluate does something better than what an average player would do, it's +1, if he does something really unexpected, it's +2, if he does something that you have seen countless of times, it's 0 (that's what an average player does, or "expected plays" so to speak), and the same way for -1 and -2. That way you can quickly recognize how worthy each play is. Your job is to recognize unexpected plays and see if they are positive unexpected plays or negative. The feel of what an average player would do in such situation comes purely from experience and watching hundreds and hundreds of games. Our brain is pretty smart and, i believe, it subconsciously caughts on what majority of players does. So what you are doing when evaluating players is continually comparing his performance with an abstraction of an average player you have in mind. Why i think that way of looking at performance is the right way it's because it gets rid of statistical bias. Tap ins lose on worth and plays that give an actual quality to the team are rewarded with some points. But again, if very good off the ball movement lead to tap in, he still gets some reward so you can still perform good scoring only tap ins. It's pretty efficient way of evaluating things imo. 2nd question: The system is designed for evaluating full performances watching full games because if you want genuine results, that's the only way to go. If your question refers to players with higher or lower work rate, i thought of it as bonus points. Someone who makes a lot of highlights was very engaged in the game and gets additional points opposed to some who was a bit disconnected and has shorter highlgihts. How much bonus points should be given is up for a debate. I was thinking of evaluating Ronaldo's performance in 2002 WC final watching full game so that you could all see the system in work and the way i genuinely thought of it, so i will probably do that in upcoming days. i believe when the system is put in practice it will start to make a lot of sense to you. It's seems to me that i showed it to be more complicated than actually is. Exactly, it's not easy at all. Making something less complicated than what i have presented so far seems pointless to me, because it won't be accurate evaluation. It's either all or nothing on this one i am afraid.
As a useful link, if you did want to go beyond Ronaldo's 2002 game: http://footballia.net/ I lose track of who knows what on the forum, but maybe you wouldn't know about that site. You can search by player name for full games or navigate around to find games from different leagues etc. Not that I suggest you should spend loads of time rating full games unless you really want to!
Great link! Too bad they dont have imho the best performance of alltime; Sinisa Mihajlovic vs Sampdoria 98.
Gianfranco Zola his best game for the national team is (by far) the 1997 qualifier against England. https://eu-football.info/_player.php?id=23522
I watched this match the past few days http://footballia.net/matches/germany-spain-world-cup France Football had only Sergi as 4/6 and the rest on 3/6 or lower (Abelardo, Luis Enrique, Salinas, Kohler, Brehme, Moller were on 2/6). Especially for the first half I'd say this is a good game for Guardiola. Had even a couple of nutmegs, took all set pieces, intercepted well and pro-actively at times. Then in the 2nd half I'm not sure, with Caminero becoming fractionally more visible (Caminero, Guardiola, Luis Enrique and Hierro the ones with most creativity, but Hierro did not really show this). Guardiola was subbed out a quarter before the end. For 2nd half I'm not sure who was the best (or least worst). Maybe Caminero, for a stretch. Maybe Matthaus for 25 minutes, who made up for a so-so first half where his timing and interceptions were off. Maybe the peripheral Effenberg at times ('peripheral' in getting offensively/defensively involved in play, seeing the ball). 2/6 for Luis Enrique and Moller looks harsh - Luis Enrique was one of the few players who could beat a man (sometimes ending with a foul). It is that he scored a goal but otherwise Klinsmann looks a better candidate to me for a 2/6, then. It isn't one of the classic matches of the tournament, no (like two or three of the quarter final matches).