Winrate is the easiest metric to get on a player and certainly the easiest to understand. But is it really a reliable measure of a players skill and if so, what other information can we glean from it? After all, in random battles you are only one player on a team of twelve and the carry potential, even on carriers, can’t solely win a match. So lets looks at some of the numbers.
As a measure of how often a player’s team wins battles by percentage, winrate should have some correlation with a players ability to have an impact on other skill metrics, such as their ability to do damage (accuracy & average damage) and their ability to survive and indeed it does. In Figure 1, the correlations between these figures and winrate are significant. This means that the probability that they are occurring by chance is very low, in this case less than .1%. As a point of note, the highest correlation for winrate is to Survival Percentage with a p value of .432, which given the factors involved and the sample size used for this analysis, is pretty good. What this means, in short, the more often that you survive in a given battle, the better your chance of winning. This makes sense doesn’t it? By surviving, you are more likely to have an effect on the outcome of any given battle. Further, the next highest worth noting, though not the numerical next highest, is average damage per battle (DmgperBattle) at .292. This is important because if we look at what is strongly correlated with survival, the highest value is with Damage at .472. Again, this makes sense as the longer you survive, the more damage you are able to do. These 3 metrics, winrate, survival and average damage, are the three best indicators of a players skill when looked at together. Quite a bit more can be drawn when assessing those three metrics but those might be a topic for another article. In the meanwhile, let me explain why Main battery accuracy isn’t as important as Damage.
Though there is not a strong (though significant) correlation between main battery accuracy and damage, the correlation between accuracy and survival is negative. This means, the higher your accuracy, the more likely you are to die. To me, this is explainable due to the simple fact that the closer you get to the enemy, the more likely your shots are to hit the target. This also usually means that you are within their secondary and or torpedo range as well as being closer to the rest of their fleet. These we usually call yolo rushes and have all seen how they end – with a quick trip back to port, hence the negative correlation. Sure, other conclusions can be drawn, but this is the most likely.
Other things to consider when looking at winrate is whether or not the players winrate is increasing as this can be a decent indicator of improvement. It’s also worth noting that, if you are to look at an individuals winrate in a specific ship, anything less than 25 to even 50 battles, depending on the skill ceiling of that ship, may not be enough to accurately assess their skill with that ship. Similarly, if you are to view a players overall winrate, I would say anything less than 1,000 to 1,500 battles will not give you a decent indicator but only show you that they are still learning (or have a second account if the WR is really high).
As discussed in a thread created by LittleWhiteMouse 1 on the NA forums, you are in control of your winrate. Sure you can boost your chances with competent division mates, but as 1/12 of your team, you can make a difference. Understanding the fundamental mechanics of the game, your ships full range of capabilities as well as your own can help you get the most out of a battle. Not sailing or acting recklessly, but rather setting up or sailing in a more calculated manner will help you survive longer thus allowing you to inflict more damage to the enemy. This leads to wins and being able to perform the acts of doing lots of damage and staying alive while doing so is an indicator of skill.
As a final note, in the figure below there is a decent correlation between AvgMainShots (average shots fired) and winrate, but most significant is the correlation between damage and Average shots fired. Simply put, the more shots you actually take, the more damage you are likely to do. Failing to use those guns, especially for destroyers, can in some cases, lead to a loss. So keep up the bombardment and remember – have fun!