All Time overrated Player´s

xG does not directly quantify or reflect shot selection (unless in the situations that a player may be a specialist at a specific difficult shot - which is actually the case with Messi), it purely reflects at the moment that the shot is taken the probability of a goal.
It in no way provides any information as to whether that shot should have been taken in the first place. An aggregate xG value does not reflect the volume of plays that ended in shots.
Shot conversion and shot volume will tell you if the player is taking an inordinate amount of shots relative to peers and how well their production is relative to that.
That data tells us that with the historical standard set by other dedicated goalscorers, Messi took an inordinate volume of shots and did not produce the goods. Meaning his performance as a goalscorer whether due to decision-making, finishing or both is below par - in fact extremely below par compared to anybody with a minimum shot volume of 20-30.

 

You are welcome to engage in film analysis of whether Messi made the right decisions to take shots in particular situations where he took shots. I’ve actually pointed out to you already that that’s an analytically separate question from finishing and specifically invited you to go forth and analyze it. If you want to make an argument on that, do it. The argument about finishing is over.

 

Like I said, as soon as I started to provide Opa's proprietary data (aka industry standard) relative to a point YOU raised and YOU made (not me, I never brought up xG) and made it the crux of your argument, you tuck tail, hide, and spew nonsense to obfuscate dialogue.
It is okay, I will provide the full references and sources and numbers.

 

Look here folks, this guy gets proven wrong demonstrably in every parameter, then runs and hides.
A list of things he was proven demonstrably wrong on with hard facts:

-Claimed David Villa was not the standout player for spain in 2010 (contrary to POTT rankings) on the basis that he likely benefitted from Xavi's midfield passing serving him up chances. Said that he likely matched his xG.

Truth: David Villa outperformed his xG by 3! And he scored a goal from near the halfway line, and another that is a dribbling solo goal wonder.

-Claimed Messi's 9.2 non-penalty conversion rate was falsified by me.

Truth: It was accurate and he never apologized for accusing me of being deceitful.

-Miscalculated Messi shot numbers by using rounded numbers from sofascore.

-Claimed multiple times that Messi outperformed his career WC xG (for the record, this is a very deceitful calculation - Messi underperformed his xG in 2022, 2018, 2010. To bring his performance from 2014 to "paper over" his finishing in those circumstances is deceitful).

Truth: Messi underperformed his xG and he has gone into a full panic as I am producing the data.

-Claimed Messi has a higher shot conversion than the WC conversion rates during which he played.

Truth: He was provided with a full panel of statistics which have put this silly argument to rest.



I would like to note that half the points he raised are strawmen - ie. intellectually incorrect on fundamental grounds. For example, whether Messi did better than the conversion rate of all shots ever taken in his WC is NOT the standard Messi should be compared to. Messi is compared to his peers - ie. other forwards who had a high proportion / volume of shots. That is the crux of the issue.
In any case, even compared to the general population of footballers - goalkeepers, midfielders, defenders, attackers, Messi performed less well to that standard. But it remains a strawman.

 

You are an unbelievably toxic human being, who takes someone not wanting to engage further with you in an unpleasant and already-played-out discussion as “run[ning] and hid[ing.” This discussion is over. I was right about finishing. That has been conclusively established by data that both of us have found. If you want to make a separate point about decision-making, be my guest. Of course, you’d have to contend with the fact that Messi’s overall conversion rate was slightly above the overall conversion rate in those World Cups (as I’ve already shown), suggesting that he was not actually taking higher-than-average difficulty shots (when we take as given that we know his finishing was roughly average compared to xG), but rather was taking roughly average difficulty shots. So that’d make your argument harder to make than it would be with someone who systematically on average takes way-harder-than-average shots (i.e. someone who would have a lower-than-average conversion rate even if they matched xG). But by all means, you can go ahead and try to make it. It’s not a point I’ve been really arguing about in any substantial way, because I was focused on the finishing issue. That discussion is over, so feel free to pivot to a point about decision-making. Given the conversion rate being better than average, though, I think you’d really need to engage in some film analysis to make a strong point on that, because it’s by no means obvious from the raw data.

 

I will be posting the data, you have no right to tell me that I cannot post this data.

 

Stop with the melodrama. And if you want to claim your argument was really about decision-making and not finishing, then I don’t understand why you refuse to actually focus on decision-making.

 

My argument was about goalscoring and shot conversion and shot volume in comparison to other dedicated goalscorers in world cup. That was exceedingly clear. You then tried to minimize the poor shot conversion, by derailing the conversation into a finishing conversation where the crux of your argument was that Messi had outperformed his xG (albeit by smaller margins than his club form). In that regard, I have proven you false and will compile a sum-total of the objective values (values that I had to force you - and myself - into recovering, since you built essays and statements on your OWN estimates and used that to level conclusions and shut down discussion).

Why are you still replying to me when you said you were done?
There's nothing to fear with what I am about to post.

 

Okay, these are just wild and misleading attacks—all of which have been addressed and refuted already in these threads but you keep making them because it’s all you have to save face. You are a miserable human being, who consistently behaves like a severely mentally ill person on BigSoccer. I’m not going to keep spending my time constantly fending off the same misleading personal attacks over and over again. Nor do I want to read them, so I’m putting you on ignore, as I should’ve done long ago.

You’re a waste of time to engage with, and a wildly unpleasant person to interact with, as others have also pointed out to you. You are way too emotionally invested in what you discuss here, and appear to be pathologically unable to admit when an argument you’ve made is wrong, and will instead lash out with an unceasing barrage of misleading personal attacks instead. For the sake of any human beings that interact with you in real life, I hope you’re just playing a deranged character on BigSoccer and aren’t like this in real life. Maybe that’s the case—I don’t know. Either way, I’m done with you. Have a good day!

 

As predicted, full mental breakdown when xG numbers are provided.

For the record, the sum-total of individuals who have personally attacked me are literal Messiphiles (Messi/Barca tags) who have a strange emotional attachment to their god.

 

Messi Goals vs. xG in World Cup for available data (excluding penalties)

2010
Goals 0
xG 1.74

2014
Goals 4
xG 2.7

2018
Goals 1
xG 1.64

2022
Goals 3
xG 3.22

Total
Goals 8
xG 9.3
(-1.3 compared to expected goals)

 

Messi's Last 4 Copa America's (for which xG data is available)
COPA 2021, 2019, 2016, 2015

8 NPG, NPXG 7.98

A reminder that he took ?13 shots in 2011 and scored 0 goals, so in his Copa Career, he underperforms xG.

 

One quick very relevant thing I just randomly found on FBref while researching something else (about Tadic: definitely see my recent thread about him—very interesting player!) that I want to note for people:

https://fbref.com/en/expected-goals-model-explained/

FBref xG data is from OPTA: In the above link, they specifically say “The model that FBref uses is provided by Opta.” And FBref’s explanation explicitly mentions how “older models” would’ve come to different results. I don’t know if the OPTA xG data provided by others in this thread has been based on the output of older OPTA xG models or just is based on a source providing incorrect information in some way, but basically the FBref xG data for WC 2018 and WC 2022 that I cited is definitionally xG data actively provided by OPTA based on an ongoing partnership between FBref and OPTA. And since those two World Cups were the World Cups where higher xG numbers purportedly from OPTA were recently cited here (and merely described as coming from a “source,” and therefore with no link), I think those higher numbers should pretty clearly be ignored, in favor of the ones I cited here, that were based on FBref data (which, again, partners with OPTA and explicitly gets its xG data from them and their most recent model).

See my analysis here that uses FBref/OPTA data for WC 2018 and WC 2022, as well as OPTA data for WC 2010 and WC 2014 (albeit only alleged OPTA data for WC 2014 as stated by another poster without any actual evidence presented), and ultimately leads to a conclusion that Messi slightly outdid his xG in the World Cup: https://www.bigsoccer.com/threads/all-time-overrated-player´s.2126873/page-35#post-41697632.

Anyways, if you’re reading this, come tell me your thoughts on Dusan Tadic here: https://www.bigsoccer.com/threads/how-good-was-dusan-tadic.2127543/!

 

Some 12H after kicking and screaming about how you were quitting the discussion and after thoroughly insulting me, you are back with this nonsense?

Buddy, the data I provided (screenshotted) is directly from OPTA.
If I wanted to inflate the xG numbers, trust me I have access to other models with HIGHER xG values. I just decided against it arbitrarily and to make sure all the data comes from the same data provider/source.

example
Other models have WC 18 at nearly 2, and WC 22 at 3.46. I generously stuck to the values that could be provided from the same data source.

 

Just so I’m clear about the importance of my post last night, here’s a summary of the bottom line:

There’s agreement that OPTA xG stats are the “industry standard.”

FBref xG data is from OPTA. Statsperform xG data is from OPTA. OPTA Analyst xG data is from OPTA.

We have Messi’s WC 2022 and WC 2018 xG from FBref. We have Messi’s WC 2014 xG from Statsperform.* We have Messi’s 2010 WC xG from OPTA Analyst. Therefore, all of the xG data I provided is from OPTA (the “industry standard”).

And that xG data showed that Messi outperformed his non-penalty xG in the World Cup overall, unless his two shots in WC 2006 were 0.46 xG or more. I provided videos of those remaining two shots (which included one good inside-the-box chance but at a bit of an angle with a set GK, and one shot from far away). While we do not have xG data for those two shots, it is IMO not particularly plausible that those chances amounted to 0.46 xG (anyone is free to watch the videos I’ve provided of those shots and see what they think). Therefore, using the industry-standard OPTA xG data, we can say with a high degree of certainty that Messi outperformed his non-penalty xG in the World Cup overall. EDIT: My post just below this provides updated info with more precise data—we now know Messi’s exact WC non-penalty xG overperformance was +1.02.

The information that was presented to refute this was purported OPTA data for WC 2022 and WC 2018 that was allegedly provided to a poster here by a vaguely-described “source,” and then screenshotted with no context (and no link). Meanwhile, FBref has an active business partnership with OPTA under which they get OPTA’s statistical data, including the output of OPTA’s latest xG model. If one wants to believe that random screenshots that a poster says they got from a “source” are a better indicator of OPTA’s latest xG data than the current public output resulting from an active business partnership with OPTA, then…okay, I guess. But I think it’s obvious to any sane person what is clearly more reliable here.
_________

* NOTE: The Statsperform xG data we have for WC 2014 is actually only a vague claim from a poster about what a “contact” told them about Statsperform data, with, as far as I’m aware, no actual evidence provided backing it up despite such evidence being requested. However, for these purposes, I’ve nevertheless taken the vaguely-provided WC 2014 number at face value. EDIT: I should note that my below post is strongly suggestive of the idea that these WC 2014 numbers are not correct, since Messi’s WC xG according to OPTA is pretty clearly too low for this number to be correct.

 

Oh, actually we can say with precise certainty exactly what Messi’s WC non-penalty xG overperformance was according to the industry-standard OPTA data, and it was +1.02.

Lionel Messi's World Cup record before tonight. He's never scored a knockout stage goal.#ARGAUS | #FIFAWorldCup | #Qatar2022 pic.twitter.com/Bj17vTTy5d

— Opta Analyst (@OptaAnalyst) December 3, 2022

This is a tweet from OPTA from just before Argentina’s match with Australia in WC 2022. It has Messi as having a total of 8.52 xG in the World Cup to that point (and based on the number of goals mentioned there, it is 100% clear these numbers include penalties). We also know from FBref data provided by OPTA that Messi went on to accrue 4.2 more xG in WC 2022 after that tweet. So that’s a grand total of 12.72 xG for Messi, with penalties. He scored 13 WC goals, so we know that, including penalties, he outperformed xG by 0.28. As has been pointed out, we also know that the xG on penalties according to OPTA is 0.79. And Messi had 6 penalties in the World Cup. Therefore, that means that, according to OPTA data, Messi’s non-penalty xG in the World Cup was 7.98. Messi scored 9 non-penalty goals in the World Cup. That means that, according to industry-standard xG data, Messi outperformed his World Cup non-penalty xG by a total of +1.02. Even more than I’d thought!

I’m sure the argument will now seamlessly move to KO-stage xG, without acknowledgment of being incorrect (and nasty in the process) here. And I won’t even get into the fact that clearly incorrect data was sketchily and aggressively being passed off as OPTA data. Ah well. Have ignored, so I won’t be able to see where the merry-go-round moves to. But just wanted to make sure everyone else has the correct information here about Messi’s WC xG overperformance.

 

This poster is not just possibly unstable, he is demonstrably unstable, erratic, and behaves uncontrollably:

Has since posted 3-4 times.

 

Now that we’ve conclusively established Messi’s exact WC xG overperformance, I think there’s some interesting follow-on questions about this, as well as some interesting notes based on OPTA data on Ronaldo that was presented.

Messi Compared to Himself

The first is about the extent to which Messi’s overall xG overperformance in the WC was disappointing compared to his norm. Messi outperformed his non-penalty xG in the World Cup by +1.02. That’s in 98 shots, and 2315 minutes. So he outperformed his xG by 0.0104 per shot, and 0.040 per 90 minutes. It’s also a 12.8% xG overperformance (though I don’t like that way of measuring xG overperformance). We have Messi’s xG overperformance data using OPTA data on FBref from 2018-2019 onwards (for league + CL). In that timeframe, he overperformed xG by 0.0318 per shot, and 0.16 per 90 minutes (note: while it’s not OPTA data, this is also pretty consistent with the 538 “Messi is Impossible” article that said that from 2010-2014, Messi outperformed xG by 0.038 per shot). This also amounted to a 28.8% xG overperformance.

So, as expected, Messi did not do as well compared to xG in the World Cup as he tended to do on average. How much below Messi’s own norm did he do? Well, if we take that typical xG per shot, xG per 90 minutes, and percent xG overperformance data and map it onto the WC shots and minutes, it’d tell us that if Messi had overperformed xG like he typically did, he’d have scored about 1.3 to 3.1 more WC goals (specific number depending on whether we adjust on a per-shot, per-90-minute, or percent basis). Which is not an insignificant underperformance compared to his own norm, especially at the top end of that range. Of course, we don’t really know the reason for this (whether it was having a different shot diet in a way that he’s worse at finishing with, just random variance, nervousness, or something else). But what seems simultaneously true is that Messi finished pretty well in the World Cup but still meaningfully less well than his normal outlier-level finishing.

Messi’s Shot Difficulty

The second is about the difficulty of his shots. Messi’s exact non-penalty xG in the WC was 7.98, and that was over 98 shots. So Messi averaged an xG per non-penalty shot of 0.0814. The overall non-penalty conversion rate in those WC’s was 8.70%. This means that Messi had a roughly average shot difficulty, but it was actually slightly harder difficulty on average. Does this tell us he was taking shots that were too difficult? Maybe. It’s hard to really draw a concrete conclusion on this though, since the average difficulty of his shots was still quite close to average. It seems like something we’d have to engage in a lot of film analysis to come to a negative conclusion about, since the wisdom of taking a shot is definitely context-dependent and the overall average difficulty of his shots was pretty close to normal.

Cristiano Ronaldo’s WC Data

One thing I’d note is that we have Cristiano Ronaldo’s total OPTA xG from a tweet that was previously posted here. That tweet was from after the 2022 WC group stages, and said Ronaldo had a 10.1 xG in 100 shots. We know from FBref that Ronaldo proceeded to accrue a total of 0.09 xG in the rest of the tournament, so he had a total of 10.19 xG in the World Cup. He took 4 penalties, so at 0.79 xG per penalty, that means he had 7.03 non-penalty xG. Ronaldo converted that to just 5 non-penalty goals, for a -2.03 non-penalty xG underperformance overall in the World Cup. If we include penalties, the xG underperformance was -2.19. So this was definitely worse finishing from Ronaldo in the World Cup. There’s also some interesting info here about shot difficulty. Ronaldo had 98 shots as well, so that means the average non-penalty xG of Ronaldo’s shots was 0.0717. That’s low enough that I’d say it’s meaningfully below the overall non-penalty conversion rate in those WCs of 8.70%—which means he was taking significantly above-average-difficulty shots on average. On one hand, that’s not surprising, since we all know Ronaldo takes some wild shots. But this more difficult shot diet suggests a much stronger inference that there was bad decision-making involved—though it could of course also in part be that Portugal is a lesser team. We’d probably need film analysis on this even for Ronaldo, but there’s at least a stronger inference from the raw data.

@Sexy Beast

 

Just assume Messi scored one of those non penalty shots like the chance vs Belgium and see if the numbers are significantly different than Messi in general.

I would say that pressure and nervousness has got better of him for some performances especially at earlier stage of his career.. there is no running from that. That probably amounted for few non penalty goals which makes the whole difference.

The more surpising thing perhaps is that Messi was not able to generate more xG which he actually excells at, but that difference is explained by having inferior team in Argentina than Barcelona at their peak.

Overall i think conclusion is pretty clear. Messi wasnt a disappointment in world cups (that could have been concluded by just watching), but also he didnt put in special kind of performances 8n world cup which you would assume he would be motivated to do.

After all, he really did care about world cup unlike random games in la liga.

So a bit of disapointment can be associated here that he didnt perform at the better end of his average but was merely in his standard.

This also i think gets down to pressure and amounts to maybe 3-4 more none penalty goals overall in his wc career.

 

For the sake of transparency and the community,
FBref was provided "a model" by OPTA and a dataset. Which model, what variables are inputed is entirely at their discretion. Different data providers use or have used different OPTA models.

So a 3 horse race in the #EPL. Time for a nerd thread looking at actual performance vs 3 XG models (Infogol, FotMob (opta) and Understat). Some absolutely remarkable things going on here pic.twitter.com/NJufgr7oIW

— Conor Bradley Fan Account (@DanKennett) October 24, 2021

To add, their partnership with OPTA started in Oct 2022. Prior data is from STATSBOMB, not OPTA.

Does this mean you will be using the Opta xG model as opposed to the statsbomb one from now ?

— Akash Baga (@akash_011) October 25, 2022

One thing that the data community agrees unanimously on is that you should never add values from different models.

Opta additionally updated their model soon after the World Cup (sometime in the spring).
The data I provided is from the same OPTA model (besides 2014 - which was from an article @lessthanjake posted - but is presumably also OPTA).

If any of this information requires referencing per your standards, I can provide it.
To be clear, people in the data community were not happy about Fbref/Opta:

Is the Opta xG model really worse than the Statsbomb one ?

— Sam (@y0Sato) October 25, 2022

opta’s xg model is not good. this isn’t new, they switched data providers earlier this season

— HH (@vxyq339) April 11, 2023

https://twitter.com/Billy_LDN_/status/1592215626751553536

An example of Opta's old xG model underestimating proabi

The new opta model post update:
https://twitter.com/TonyElHabr/status/1695923553659449549

https://twitter.com/SebWassell/status/1584987496324046848

Messi's Last 4 Copa America's (for which xG data is available)
COPA 2021, 2019, 2016, 2015

8 NPG, NPXG 7.98

A reminder that he took ?13 shots in 2011 and scored 0 goals, so in his Copa Career, he underperforms xG.

So his xG underperformance is a fact of his entire international tournament career.

 

For example there is no way Messi's 2010 xG amounts to 1.7?

 

Yeah, generating xG is a big deal, but it really is a team effort. Messi is absolutely elite at advancing the ball to more dangerous areas, but if two teams are playing conservatively (i.e. almost all the time in the World Cup) and the other team has an organized defense (which at that level they typically do), it’s extremely hard to go that extra little bit to get from a bit outside the penalty area (i.e. a spot where the xG on a shot wouldn’t be high) to well into the penalty area (i.e. a spot where xG would be high). It’s not particularly realistic for one player to do that with pure individual brilliance against an organized modern defense of WC-level players when your team and the opponent’s are playing pretty conservative tactics (there’s just no space and the defenders are too good). In order to get those sorts of chances, you need your team to do stuff. Whether that’s people creating havoc on the wing or playing great one-twos to open up play in the box, or something else. Argentina hasn’t lacked the quality on paper to do that, but we haven’t exactly seen it in reality—whether that’s due to certain players’ form or just the inevitable result of conservative tactics and good opposing defenders, I don’t know.

I think this is the sort of place this discussion should be though. We now know conclusively that Messi overperformed his WC xG. It wasn’t by as much as normal for him, but not performing to his own outlier-finishing standards isn’t exactly a convincing overall criticism. To the extent he might be due real criticism, I think it’d be on this actual question of generating xG in the first place. To what extent was it his fault that his total WC xG wasn’t particularly high? Was he too quick to shoot and was ending attacks that might’ve generated better-quality chances? Was he himself lacking compared to others in quality of dribbling or short passes in tight spaces that can open up stout defenses? Was he not making off-ball runs he should’ve been making? Or was his team just playing conservative tactics that, essentially by design, chose defensive solidity over the team being able to really generate high-quality chances? Some of this stuff is likely not right (I don’t think Messi was lacking in ability in tight spaces in these tournaments, compared to any reasonable baseline). Meanwhile, some of these would take film analysis to really drill down into (like if he was shooting when he shouldn’t have). I do actually remember thinking at the time that a lack of activity hurt his ability to get chances in WC 2014 (where I think, to be fair, he didn’t seem entirely fit—with the throwing up on the field and whatnot), but it’s not like he got way higher xG in other tournaments, so I don’t know how determinative that was (or even if my eye test on that was actually correct).

Anyways, that’s where a discussion on Messi’s WC goalscorer should be. We know from the data that his finishing in the World Cup was pretty good (albeit not quite at his typical outlier standards). That’s obviously not a debatable premise at this point. So the only question regarding Messi criticism in terms of WC goalscoring is just to what extent, if any, he can be criticized for not having a higher xG.

 

Have been doing some searching through various football stats twitter accounts to find interesting info (have found a lot of cool stuff about random things that I plan to note soon in various threads as they become relevant!).

One thing I saw that is relevant to some of the discussion in this thread:

We used the “industry-standard” OPTA to measure Messi’s WC xG (recall: the precise data we have comes from an OPTA Analyst tweet from December 2022 that provided full OPTA xG data for Messi through the WC 2022 group stages, and then we looked at FBref’s OPTA data for the WC 2022 KO stages to get the rest, so all of the data used is from OPTA). Of course, OPTA is the “industry standard” (and perhaps even moreso after their update to their model around March 2022, before they reported the data our conclusion is based on–see description of that March 2022 model change here: https://www.fantasyfootballscout.co...s-to-optas-expected-goals-xg-stats-explained/), so providing OPTA data was appropriately the end of the discussion on this issue. That said, there are of course other xG models, and I did just happen to come across relevant data from the Statsbomb model that I believe is at least pretty well thought of. Specifically, while I am not aware of them ever providing full WC xG data for Messi, they did provide WC 2022 xG data for Messi:

Lionel Messi: World Cup Winner pic.twitter.com/9zt9l8lLTD

— StatsBomb (@StatsBomb) December 19, 2022

That has Messi at slightly below 0.08 xG per shot. It’s clearly non-penalty shots (since it’s not mathematically possible to be that low if it were counting penalties). Messi took 27 non-penalty shots in that World Cup, so that means that Statsbomb’s xG model had Messi somewhere a bit below 2.16 non-penalty xG in WC 2022. This is particularly interesting, since it’s actually notably lower than the FBref OPTA data we have for that World Cup, which was 2.7 non-penalty xG in WC 2022. So, while we don’t have Statsbomb xG data for all of Messi’s World Cups, the data we do have suggests that it might tend to estimate lower xG for Messi than the OPTA data we have (and that it definitely did so by a meaningful amount in WC 2022). And, in that case Messi’s xG overperformance would be even larger under that model. Of course, we can’t say anything at all for certain about this, since we only have one World Cup’s worth of data from Statsbomb, but what we have is certainly suggestive of Statsbomb’s model actually being more charitable to Messi than the industry-standard OPTA data that already has Messi at +1.02 xG overperformance! It’s a minor thing, since obviously we have all acknowledged that OPTA is the industry standard, but I figured I’d note having found additional info that others might find at least a little interesting.

 

I hope everybody has had a good laugh at this spin.
Please do remember the words "what we have is certainly suggestive of Statsbomb’s model actually being more charitable to Messi than the industry-standard OPTA data that already has Messi at +1.02 xG overperformance! "

Tomorrow, you will be provided Opta data distributed by SkySports, a clear explanation for why older values from Opta are unreliable, and a clear evidence-based demonstration that the Statsbomb does not in fact "underestimate" values relative to opta ie. variations between xG values does not suggest that a model will consistently produce lower or higher values on the whole.
Finally, the suggestion that 29 shots in WC2010 amounts to some 1.7 xG is an insult to everybody's collective intelligence.

We will also proceed with a valuable discussion on Messi's finishing performance over a larger sample size - Copa America and World Cup. This way we will achieve the full gamut of analysis of his abilities in international cups. This will help us ask the tough questions - why does Messi's finishing performance drastically take a hit in the most important matches of his career?

 

So what did he say wrong? Where has he made the mistake? Everything checks out and he provided information and sources.

Stop being an idiot and stop ridiculing every post that you disagree with.

All you do is hype up some unkown sources and then write up some numbers in thread.

I, in fact, do have proven and reliable sources of premium opta package that goes into 4th decimal point that all the best managers use in the world and they say following:

Messi wc overall:

3,8732 xNPG

Yes. That is including all 5 wcs.

Messi actually is the best finisher in hsitory of wc.