Vilarrubi
New member
BBZ, I've seen you bring up these stats about 100 times both to praise Rakitic' god-like defending skills and to criticize Arthur's. While I can understand the latter - Arthur is indeed no defending machine (though I'd take him above Rakitic any day) - the former is just a result of wrong reasoning.
You're using 4 stats here to supposedly prove Rakitic has the best defensive stats out of all of our central midfielders. However, there is only stat pointing this way that actually has some (statistical) importance to it. Allow me to explain:
- Interceptions: 1.5 is not bad, in line with Busi so you could say he's the joint best in this area. Sure, I'll give you this one. Let's look at the other stats.
- Blocks: even mentioning this stat is absolutely ridiculous. You're talking about a stat where Rakitic is averaging 0.2/game. That makes the stat completely irrelevant for two reasons: Firstly, it is statistically insignificant. You don't compare metrics when the number of observations is this low, it just doesn't make any sense. If in the future Busi were to average 1 extra "lucky" block every 5 games over Rakitic (which could very well happen, after all what's one block every 5 games), he would end up making 0.2 blocks per game more than Raki. According to your logic of simply comparing stats to determine who's best in a skill, Busi would suddenly be the better "blocker". Do you think that makes sense? In more statistical terms, for a metric with such a low number of observations, randomness plays way too big of a role to make any interpretations. I'd recommend you do some research on the law of big numbers if you don't understand. Secondly, you have to think about the footballing significance of this stat. Sure, Rakitic' averages 0.2 blocks per game more than Arthur, that's 1 every 5 games. Is that something we actually value? Is one block every 5 games something that's even worth mentioning? I'd say hell no! No statistical significance + no footballing value = dogshit stat
- Clearances: same reasoning applies, to a lesser extent (I'll give you that). The difference with Busi and Vidal is really small however. Again, the difference is one extra clearance every five games (compared to Busi and Vidal). Is that even worth mentioning? Your call.
- Tackles: Aaah, the stat with the most statistical significance. This stat is much less skewed by random events (e.g. lucky tackles) than the other metrics. And what do we see? Rakitic is not the best one here, not even nearly. In fact, Busquets averages almost 4x more tackles every game than Raki! In other terms, he makes 2.5 more succesful tackles every game. Finally something that actually has some value for a football team: I think we can all agree that 2.5 tackles in a game is actually useful for a team.
So to sum up, you used 4 stats to prove your argument, of which
- 1 stat is relevant
- 1 stat is statistically completely insignificant with no footballing meaning/value, and in which Rakitic is joint best with 2 of the other 3 players
- 1 stat which is statistically fairly insignificant with no footballing meaning/value
- 1 stat that is statistically significant, that has footballing meaning/value and in which Raki gets destroyed by 2 of the other 3 players
Hope I made my point.
Thank you for reading, can't believe I wrote this whole BBZ of a bible...
I'll make sure to copy this monster so I can reply the same thing every time you bring up these nonsense stats.
Peace out, and make sure you respect statistics kids
When the stats guy gets out-statted.