Jump to content
tc1259

Stats - what do they mean? what do they mean to you and why?

92 comments in this topic

Recommended Posts

Moderator
590 posts
2,995 battles

Background: We have a thread about stat bashing, but I thought I would separate this conversation out in to it's own thread. I briefly touched on a number of the stat point's but I think i'm going to expand it out a little more here. 

Now to the original conversation, where I will expand out.  I'm a bit of a stat nerd, after all it's what I did for a paid professional job.... here is my take on the stats that are collected

Additionally and I feel that this is important, as it's usually always confused - Stat's and Game Play Styles - they are mutually exclusive to one another, this conversation isn't about Game Play style, it's looking purely at the stat and what I feel by looking at the data presented is more indicative of a players skill level.

Let's start with the one that's always brought up. Win Rate

Win Rate: Win Rate in world of warships, at least in random and ranked is kinda hard to pin down as the most important stat, the reason why is the following. You might be a great player, but if everyone around you in your team isn't. Then no matter what you do, the % chance that you'll win the battle is quite low. Now Yes, one can argue that over time this shouldn't really be a factor, but remember - over time doesn't actually count. It's the battle your in that gives you the win or loss. 

I have seen that the argument of 'when i've done this' in a match, we've usually won.... that's fantastic, congrats. But in any battle there are pivotal moments, these are X factors and some people can see them, others cannot.

By in large, an individual's play cannot statistically influence a battle, there are other factors involved -  but occasionally that can happen. As an example - you might yolo at the perfect moment, distract the enemies flanks, who then open themselves up to being attacked, causing them to take further damage and ultimately winning your battle. 

But this wont always happen in every match, every time... 

So what are we looking at here in Win Rate: well, taken mathematically it's this simple. You are 1/24th of the combination that can either win or loose the game, additionally you need to factor in RNG. So really your 1 in 25. 

Additionally to this you now need to work, without knowledge of who your team mates are there skills and weakness's and ultimately x factors which are not quantifiable to work out that from the moment you enter the game your likely hood of winning this match is actually quite low. No matter if your the best player in the world, this will only increase your winning % and not make it a guaranteed thing.

Overall, whilst win rate can be used as a guide, it's not accurate, nor is it really comparable to others due to the nature that your not always with the same players. This is also why you see a difference between divisioned players and non. You have two additional players in your division, you do increase the odd's of a win, but at the end of the day the Win / Loss stat is extremely variable.

Again, pure stats, nothing to do with X Factors and un-quantifiable elements


Average Damage: Unlike the above stat, this one I feel is actually the most important stat you can have. And I know that will be a bit of a controversial view on things, but I'll explain why. 
 

  1. This stat is the only stat you have direct control over
     
  2. This stat is the only stat that can have a direct relationship to your team winning / loosing a battle - now don't get me wrong, we've all had games where we've done 100k plus damage to loose, but if you look at your team in those losses, you'll see there average damage was quite low
     
  3. The more damage you do, the less HP the enemy has - the greater your influence to win the battle becomes.  

Personally, if i'm looking at a player who has say a 40% win rate, but has a much higher average damage than say some other players - I'll generally take them over someone who has a higher win rate - why, cause I know there doing everything they can in the situation there in. They are doing damage, obviously more information is needed at this point, but this is going outside of the stat's and more in to tangible play style / personality. 

Average damage to me says as an example. If the Average damage on the server is 90k for say Montana, and I'm doing 180k - I'm pretty sure that the player has a real handle on that ship, knows what there doing is getting them results and lets face it, they are now 'killing' a large portion of the enemies hp, that in some cases could be a couple of ships.

Please be aware, that I am aware that some ships have low damage output, when I talk average damage, it has to be seen in layers - ie: Overall and at ship level. 

Battles: Another Meh stat, really this doesn't count for much - other than Experience at playing the game, we all know it takes a while to become good at a ship. 

Average Experience: Another good indicator, the higher the average exp, the more work the player is doing in the battle. This generally is a direct result on the damage / in game actions. If from a stat point of view I'm looking at Avg Damage + Avg Experience + Win rate and it's telling me a good story, then i'll be saying that hey, the player in question actually knows what he's doing. 

Survival: Meh - you can survive a lot of games by sitting in the back corner not doing much - but if this is higher than average, as well as Avg Damage and Avg Exp, then this player is starting to show that they are a good player

The following two stat's I don't overly worry about - reason why - End of the Day, killing a ship whilst important, is not as important as damage. Kill Stealing is I couldn't care less, working in a division as long as that ship goes bye bye. And plane deaths, well okay - if your in a ship that has little AA, why would that matter? unless your in a cv

Avg Kills
Avg Plane Kills

Kills / Deaths: 
This stat is and isn't important, depending on what your looking for, this stat is subjective dependent on if your skilled at supporting your team or being the spearhead. Individually, the top players on the asia server sit around the 2.0 to 3.7 kill to death ratio, showing there is a link there between being good as an individual and stats. But from a purely team play point of view, this kill to death drops to around 1.5 in Team Battles.

Average Tier: Kinda important, a player can have great stat's but if there simply playing in Tier 3 all the time - then there probably not going to make the transition. Personally most of my games are in 5 / 6 and 7 - so for me that's where I make money. 

On the Asia server, it currently looks like the average tier being played is 4.6.

Anyways - just remember that win loss isn't the best guide on a player, it does have some correlation - but when you factor in other elements, it's entirely possible that a decent player can have a bad W / L - 

In my opinion, people should focus on what they can control and not what they can't. Oh and remember, statistics do not matter to the individual! You only have to look at the sporting and medical fields to prove that.

I would love to see this come up as a discussion with other players opinions, but as a reminder - No Stat Bashing and just because someone may disagree with your opinion on why a particular stat is or is not important, it doesn't mean they are right or wrong. 

This thread is actually about what the different data point's mean to you as players, not how you got the stat - As mentioned above, X factors cannot be recorded

Share this post


Link to post
Share on other sites
[MEGA]
Member
4,560 posts
17,542 battles
Just now, RalphTheTheatreCat said:

HoYp1cA.gif

is that a visual representation of you gidding gud and your yellow stats becoming green?

Share this post


Link to post
Share on other sites
[MEGA]
Member
4,560 posts
17,542 battles
4 minutes ago, RalphTheTheatreCat said:

Ill be perpetually piss yellow my friend.

while im more red the the communist nation i come from

IM AS RED AS I AM

Edited by drakon233

Share this post


Link to post
Share on other sites
Super Tester
1,177 posts
4,099 battles

In my opinion, percentage damage inflicted to the enemy is a more useful metric than raw average damage, when it comes to gauging how well a player contributes to a team's victory.

I would consider dealing 85% HP damage to an enemy Gearing more "useful" than dealing 20% HP damage to an enemy Yamato, even though when it comes to raw damage numbers, the 20% HP damage done to the Yamato would be significantly larger. The reasoning behind this logic is that dealing 85% HP damage to the Gearing puts him out of action much faster - the Gearing now has to play super passively and carefully, or risk being knocked out of the game; on the other hand, the 20% HP damage you've done to the Yamato doesn't really mean much at all, and doubly-so if it's merely HE/fire damage that can be repaired back. Not to mention, destroyers have more of an influence over battle outcome compared to battleships, meaning that prioritising enemy destroyer targets increases the likelihood of winning the game.

Unfortunately, WG's stats API does not record (or rather, publicly share) percentage damage, and only raw damage, meaning that stats websites don't display such statistics either.

Share this post


Link to post
Share on other sites
Member
455 posts
8,038 battles
1 hour ago, tc1259 said:

 


So what are we looking at here in Win Rate: well, taken mathematically it's this simple. You are 1/24th of the combination that can either win or loose the game, additionally you need to factor in RNG. So really your 1 in 25. 

One big oversight here. You are assuming each player in the match has equal influence. Reality is that most of the allied and enemy players will be average or plain bad, reducing their influence to merely taking and doing damage. Have you ever seen how a good dd player can kill the enemy dd contesting the cap and proceed to make the entire enemy flank collapse and run? Or how a good DM player can deny two caps with his radar usage while discouraging enemy bbs from pushing with his constant DPM? Thats the kind of influence a good player can  have. Individual plays can turn a game around by removing key ships, denying key areas and being in the right position to exploit enemy mistakes. So a great player can have far more influence than the 1/25 you mathematically calculated. Otherwise, how come the best players on this server can hit 60% solo winrate in high tiers? How can I with my average skills hit 55% solo winrate in high tiers?
Winrate isnt an uncontrollable stat unlike what you claim. It depends on your own skills and how you play. Average damage on the other hand can easily be farmed by spamming HE at BBs while kiting away. 150k avg damage in a zao could just mean you spent 20 minutes burning a yamato. That wont win you the game though. An 80k avg Des Moines would have much more influence on the outcome of the game than the said zao by denying caps and killing dds

Share this post


Link to post
Share on other sites
Member
1,386 posts
8,216 battles

I care about WR more than anything else, but I never had the courage to ask unicums for division to boost my WR…

But I would like to point out a few details

1. The average of only a few games does not count. Those stats are not accurate enough due to the small sample size.

2. Average damage should be viewed with ship tier&type or even specific ship in mind.

3. Average EXP, IDKY takes prem acc (but no other type of multiplier) into the calculation.

Share this post


Link to post
Share on other sites
Moderator
590 posts
2,995 battles
46 minutes ago, Adm_Kunkka said:

One big oversight here. You are assuming each player in the match has equal influence. Reality is that most of the allied and enemy players will be average or plain bad, reducing their influence to merely taking and doing damage. Have you ever seen how a good dd player can kill the enemy dd contesting the cap and proceed to make the entire enemy flank collapse and run? Or how a good DM player can deny two caps with his radar usage while discouraging enemy bbs from pushing with his constant DPM? Thats the kind of influence a good player can  have. Individual plays can turn a game around by removing key ships, denying key areas and being in the right position to exploit enemy mistakes. So a great player can have far more influence than the 1/25 you mathematically calculated. Otherwise, how come the best players on this server can hit 60% solo winrate in high tiers? How can I with my average skills hit 55% solo winrate in high tiers?
Winrate isnt an uncontrollable stat unlike what you claim. It depends on your own skills and how you play. Average damage on the other hand can easily be farmed by spamming HE at BBs while kiting away. 150k avg damage in a zao could just mean you spent 20 minutes burning a yamato. That wont win you the game though. An 80k avg Des Moines would have much more influence on the outcome of the game than the said zao by denying caps and killing dds

Nope - not an oversight - I am not factoring in Play Styles or X factors as I call them, you cannot quantify the value that a person's leadership, ping, luck and experience bring to the game.  

This conversation is about pure numbers :) And I made that very very clear at the start.
 

Share this post


Link to post
Share on other sites
Moderator
590 posts
2,995 battles
56 minutes ago, benlisquare said:

In my opinion, percentage damage inflicted to the enemy is a more useful metric than raw average damage, when it comes to gauging how well a player contributes to a team's victory.

I would consider dealing 85% HP damage to an enemy Gearing more "useful" than dealing 20% HP damage to an enemy Yamato, even though when it comes to raw damage numbers, the 20% HP damage done to the Yamato would be significantly larger. The reasoning behind this logic is that dealing 85% HP damage to the Gearing puts him out of action much faster - the Gearing now has to play super passively and carefully, or risk being knocked out of the game; on the other hand, the 20% HP damage you've done to the Yamato doesn't really mean much at all, and doubly-so if it's merely HE/fire damage that can be repaired back. Not to mention, destroyers have more of an influence over battle outcome compared to battleships, meaning that prioritising enemy destroyer targets increases the likelihood of winning the game.

Unfortunately, WG's stats API does not record (or rather, publicly share) percentage damage, and only raw damage, meaning that stats websites don't display such statistics either.

percentage damage inflicted to the enemy 

Now that is a stat I would love to see, because it would indicate quite strongly how the player is handling the battle

Share this post


Link to post
Share on other sites
Member
2,214 posts
13,673 battles

A good example of why average damage is bad can often be seen late match with cruisers capping.

A good player will stop firing and go undetected until the cap is secured, then start firing. Meanwhile the bad player next to them will keep firing and not contribute to the cap because they keep getting reset. Had they stopped firing the team could cap in half the time then we could both fire.

To elaborate.

With 1 min to cap and 1 person is firing but getting reset and 1 person not firing. That's 1min x DPM and 1min x CAP.

With both not firing that's 2 x 30sec CAP and 2 x 30sec DPM. So you can still output the same DPM but you can cap 30sec faster.

Meanwhile in the first example the player who kept firing will have higher avg damage than the first player, but it was at the cost of 30sec cap for the whole team. And lets not forget you kept the other player locked down for an extra 30sec they couldn't reposition.

Edited by keskparane

Share this post


Link to post
Share on other sites
[MEGA]
Member
4,560 posts
17,542 battles
6 minutes ago, tc1259 said:

percentage damage inflicted to the enemy 

Now that is a stat I would love to see, because it would indicate quite strongly how the player is handling the battle

yes please, thiswould be one of the most useful stats to be able to see, it would go a LONG way into interpenetrating a player;s actions

Share this post


Link to post
Share on other sites
Member
159 posts
5,676 battles

To answer the question posed in the title, they mean very little to me, I play the game for fun. As long as I feel I've had a good game and contributed to the team then I'm happy. I have enough percentages and stats to worry about in the real world, I certainly dont need any more in WoWS!

Share this post


Link to post
Share on other sites
Member
980 posts
12,617 battles

Regarding win rate, personally this is how I think of it. This probably applies to most players. 

You play 10 games, or take a random sample of 10 games over a player's career.  

In 4 of those games, your team will win regardless of what you do due to factors outside of your control (top tier, noob enemies). 

In another 4 of those games, you team will lose regardless of what you do due to factors outside of your control (crap teammates, uptiered, unfavourable matchmaking). 

The remaining two games can swing either way using your own ability. That's the difference between a 40% win rate and a 60% win rate. 

So out of ten games, if I play 2 games where I can comfortably say to myself I made a key play that changed the result of the games, I will be happy. I am playing well. 

For those who have a sub-40% win rate, I have no idea what you are doing, you must be sabotaging your team and making extraordinary plays to make your team lose. 

Edited by LordTyphoon

Share this post


Link to post
Share on other sites
Member
2,214 posts
13,673 battles
24 minutes ago, LordTyphoon said:

Regarding win rate, personally this is how I think of it. This probably applies to most players. 

You play 10 games, or take a random sample of 10 games over a player's career.  

In 4 of those games, your team will win regardless of what you do due to factors outside of your control (top tier, noob enemies). 

In another 4 of those games, you team will lose regardless of what you do due to factors outside of your control (crap teammates, uptiered, unfavourable matchmaking). 

The remaining two games can swing either way using your own ability. That's the difference between a 40% win rate and a 60% win rate. 

So out of ten games, if I play 2 games where I can comfortably say to myself I made a key play that changed the result of the games, I will be happy. I am playing well. 

For those who have a sub-40% win rate, I have no idea what you are doing, you must be sabotaging your team and making extraordinary plays to make your team lose. 

That's all fine and dandy if you knew which 2 games were the specific ones you needed to contribute to. Otherwise there's something like 80% chance you made no impact.

Share this post


Link to post
Share on other sites
Member
980 posts
12,617 battles
5 minutes ago, keskparane said:

That's all fine and dandy if you knew which 2 games were the specific ones you needed to contribute to. Otherwise there's something like 80% chance you made no impact.

Well why can't you tell specifically which games they were? They'd be the ones such as where:

- You dev-striked anything at any point in the game. 

- You obliterated a enemy DD with a torpedo early on in the game which led to an entire enemy flank having to retreat. 

- You smoked up friendly cruisers (especially ones with radar), which enables them to shoot at enemy DDs. 

- You made a key play to prevent the death or deaths of teammates (especially cruisers) to enable them to keep DPSing. 

Notice how most examples involve a DD (psst that how you influence a game). 

My personal favourite - take a battleship (think HMS Hood), kite away from 1/2 the enemy fleet, who spend the entire game shooting at you, but do not kill you due to outstanding use of WASD, heal, and damage control. Meanwhile the rest of your team are free to take all the caps. 

Share this post


Link to post
Share on other sites
Member
2,214 posts
13,673 battles
48 minutes ago, blauflamme22 said:

To answer the question posed in the title, they mean very little to me, I play the game for fun. As long as I feel I've had a good game and contributed to the team then I'm happy. I have enough percentages and stats to worry about in the real world, I certainly dont need any more in WoWS!

And that is good. Stat's just often prove useful in highlighting how our feelings and reality don't align. Perhaps that is why there is a saying "Ignorance is bliss"?

I don't care if there is a bad player who is trying to improve. What I dislike is a play who subjectively thinks hes great, meanwhile ignoring the statistics that say he isn't meeting the objective, who goes around trying to tell other people how to play the game and insulting them.

Share this post


Link to post
Share on other sites
Member
2,214 posts
13,673 battles
Just now, LordTyphoon said:

Well why can't you tell specifically which games they were? They'd be the ones such as where:

- You dev-striked anything at any point in the game. 

- You obliterated a enemy DD with a torpedo early on in the game which led to an entire enemy flank having to retreat. 

- You smoked up friendly cruisers (especially ones with radar), which enables them to shoot at enemy DDs. 

- You made a key play to prevent the death or deaths of teammates (especially cruisers) to enable them to keep DPSing. 

Notice how most examples involve a DD (psst that how you influence a game). 

My personal favourite - take a battleship (think HMS Hood), kite away from 1/2 the enemy fleet, who spend the entire game shooting at you, but do not kill you due to outstanding use of WASD, heal, and damage control. Meanwhile the rest of your team are free to take all the caps. 

But you said...

12 minutes ago, keskparane said:

You play 10 games, or take a random sample of 10 games over a player's career.  

In 4 of those games, your team will win regardless of what you do due to factors outside of your control (top tier, noob enemies). 

In another 4 of those games, you team will lose regardless of what you do due to factors outside of your control (crap teammates, uptiered, unfavourable matchmaking).

So that only leaves 2 games which you can influence. Which 2?

Share this post


Link to post
Share on other sites
Member
455 posts
8,038 battles
6 minutes ago, keskparane said:

So that only leaves 2 games which you can influence. Which 2?

He didn't mean you can only influence 2 games. He meant that your influence can only hard carry 2 out of 10 games and the rest will be wins or losses inspite of your influence.

Share this post


Link to post
Share on other sites
Beta Tester
636 posts
3,472 battles

WR Arent important! I mean you know you going to end up banged in a losing or inexperience team incapable of coordination. Last time mine was 45 ( because of those noob teams ) now i currently 52 ( because of struggle on stupid teams ).

Currently now I'm just staying in-between 50.

 

You just look at Battles instead. Battles are crucial for those who are experienced in Bote games. Like example ,

0 - 900 : A Meh to levels.

+ 1000 - 3000 : Meaning that you known it quite awhile ( unless if you are Alpha or CBT , you don't need to overlooked it )

+ 3001 - 6000 : Meaning you know how to bote.

 

But overall even though you are such a good Stat player, you still sucks anyway. Even to Elite Clan players.

Its a matter HOW you Enjoying it. Or some bote that you are Superior of.

Edited by LawrenceXVIII

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×