Jump to content
You need to play a total of 5 battles to post in this section.
Spartias

Stat Tracking Self-Improvement

29 comments in this topic

Recommended Posts

393
[5D]
Members
1,287 posts
7,597 battles

I've seen the question "What is the Greatest Stat?" asked quite a lot recently. I have my own opinions about that, and I've even voiced those opinions. (Hint: it's overall win rate with reasonable asterisks attached to the theory.)

 

But my opinions on the greatest stat isn't really what this is about. This is about giving players a way to discover and manipulate data themselves. I spend a lot of time delving into the API that Wargaming so kindly provides to us, and I like to think that gives me a certain perspective on things. This spreadsheet is designed to give that ability to you as well. With the upcoming third season of Clan Battles on our doorstep, I was hoping that this could help players understand a little better where they need to prepare themselves.

 

The first thing one does is turn the sheet on, which is done in cell B1 by entering any clan tag. It doesn't matter if you're in that clan, as entering a clan tag simply turns the sheet on. You'll then be able to either type a name into cell C5 (make sure you spell it right) or select one of the members of that clan from the drop down menu there. You don't have to search for a player in that clan, but it does make it easier.

 

SPQ-Name.JPG.c4c5aa014a02d6663e6e1f902d50ead0.JPG

 

From there the sheet will begin loading data. What you're able to manipulate is in the middle of the screen labeled "Stat Goals."

 

statgoals.JPG.20f1d971bde732558fcf155728d9e5e0.JPG

 

The sheet comes preset with the above stat goals. I've so far spent two seasons in clan wars at the Typhoon/Hurricane level. When I saw a player make an influential play, or just in general really helped their team, I took notice. I have based these stat goals on those players. Basically, if you want to play at the Typhoon or higher level, these are what you'll need to shoot for. My own clan uses an adapted version of this, and they disagree with me on the numbers. Disagreement is fine! We have a few players that successfully got their flags that theoretically wouldn't make it based on these numbers. The preset values are simply based upon my own observations. And I definitely observed that there is a wholesome correlation between a player's random battles statistics and their ability to perform in Clan Battles. Once you have your own copy of the sheet however, you can change these to whatever you like.

 

These stat goals are quite important. For each one that you meet on a ship, you gain one "Qualifying Rank." Once you have met at least 4, you earn the basic descriptor "Qualified" for that ship. Exceeding the goals results in earning better descriptors on a ship. Working to earn better descriptors on a ship is kindof the entire point of the sheet. As you improve on a ship, you earn better descriptors! The sheet will not give you 4 qualifying ranks until you have at least 25 battles on a ship.

 

SpartiasDescriptors.JPG.1525cca32eab36400665045147b07552.JPG

These for example, are my descriptors for tier 10. I know... I sucketh quite greatly with the Gearing... 

 

But what about outlier ships? Ships that don't really fit the traditional dd, ca, bb, cv mold? By scrolling down below everything else, you'll see the section labeled "Individualized Stat Goals." If you set a ships' stat goals here, they'll override the more generalized versions above. The sheet comes preset with the Khaba already having individualized stat goals.

 

 

WEIGHTED STAT GOALS

Here's the kicker though. By looking below all of the individual ship results, you'll see a couple of nondescript areas (cells D80 and D82) labeled "Weighted Stat Goals" and "Weighted Player Averages." What if a player only has 2 battles on a ship, but they have 100% win rate etc? Should that be weighted as much as a ship that they have 100's of battles in? In this section the sheet takes all of the different ships, their goals, and weights them with their number of battles as the control. 

WeightedStatGoals.thumb.JPG.7beadfac433d069a195898291b6e535d.JPG

 

Those are my weighted stat goals. As you can see, I need to play more at t10, as my weighted stat goal for battles is in the red. I on average have too many t10's for how much I actually play at t10. However when I do play at t10, my other stats are all in the green. This section is best used at a quick glance to see if the player is overall actually performing at the level they should be or not. I can't take credit for this section though, the guys in OPG took the time to critique the sheet and they suggested it. They even took the time to explain the math to me!

 

Achievements.JPG.94510591e6b59e2dfad7eb211d4ba5de.JPG

 

The sheet will also take the time to look up a series of achievements for a given player so that at a glance the user may see if they have any experience in Clan Battles. All it's really saying here is that they played in those leagues, but if I recall correctly you only have to play a few battles in a league to earn the achievement from Wargaming. You don't have to earn a Stalingrad flag to earn the Typhoon league Achievement. But it does serve as a quick reference on a new player to your group.

 

 

 

Remember, this is simply a tool. However it is a tool designed for the user to set their own goals and rewards. I hope this is useful to players!

 

If you would like a copy of this spreadsheet, click here.

 

I have spent a lot of time looking for critiques on this latest piece of work. However it cannot ever truly be perfect. If you notice something is wrong, or something that could be improved, please let me know.

 

 

Updates:

 

8-14-2018 - 

     Auto-Import Stat Goals --- If this feature is turned on, the sheet will import the server averages for ships in order to dynamically adapt the stat goals to their appropriate levels. The user may increase or decrease the an added percentage above the server average by adjusting a variable directly beneath this option. It is located in the middle of the sheet below the original stat goals. The imported/dynamically adjusted stat goals may be viewed by scrolling down. If the user sets the multiplier variable to zero, one may use this to see the non-adjusted server averages/expected variables from wows-numbers.

 

     Player Rating --- The sheet now imports a brief overview of a player's "Player Rating" from the site wows-numbers.com. It also provides a link to the source page for ease of access. Individual ship player ratings are also calculated in the background using wows-numbers server averages/expected values. 

 

     Spotting Damage --- Spotting damage has now been included as a imported statistic. It is not included as one of the stats required as a qualifying rank, however the data is now available for users to use as they see fit.

 

     Potential Damage --- Potential damage has now been included as a imported statistic. It is not included as one of the stats required as a qualifying rank, however the data is now available for users to use as they see fit.

 

     Ship Name/ID Import --- This feature is an unseen feature. The previous iteration of it no longer worked as a standalone. There have been too many additions of ships to the game. This has been updated to have double the available space. This means the load times of ship names and id's has gone up slightly, however it should mean that the sheet has been future proofed considerably.

 

12-10-2018 - 

     Secondary Goals --- Sheet now allows for two separate sets of goals, competitive and secondary. Secondary goals are by default set as 15% below the competitive goals. 

     Player Impact --- Located on the top right of the screen is now a new rating. Player battle count is compared to the number and frequency of player achievements. Achievements are weighted depending upon their difficulty in obtaining and how much it generally represents impact upon the game. Generates a statistic on player impact as well as shows if the impact is over or under their win rate. This statistic isn't to be taken too seriously, as it is an experiment and will likely be tweaked in the future.

Edited by Spartias
Two new features and code cleanup.
  • Cool 7
  • Boring 1

Share this post


Link to post
Share on other sites
1,086
[OPG]
Beta Testers, In AlfaTesters
2,924 posts
10,485 battles

TLDR OP made his own version of wtr 

  • Funny 1

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
Just now, Komrade_Rylo said:

TLDR OP made his own version of wtr 

 

Heh yeah you could say that. A user controlled and user adaptable version yes.

Share this post


Link to post
Share on other sites
1,013
[OPG]
[OPG]
Beta Testers
1,451 posts
10,574 battles

My name is DOCTOR_CITADEL and I endorse this message. 

@Spartias has delivered, once again, yet another top-notch, high-utility, google sheet. 

It would be worth your time to check it out!

Trust me, I'm a doctor. 

  • Cool 2

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
23 minutes ago, DOCTOR_CITADEL said:

My name is DOCTOR_CITADEL and I endorse this message. 

@Spartias has delivered, once again, yet another top-notch, high-utility, google sheet. 

It would be worth your time to check it out!

Trust me, I'm a doctor. 

 

Wow, thank you very much for the kind words!

Share this post


Link to post
Share on other sites
1,766
[OO7]
Members
3,297 posts
9,740 battles
34 minutes ago, DOCTOR_CITADEL said:

My name is DOCTOR_CITADEL and I endorse this message. 

@Spartias has delivered, once again, yet another top-notch, high-utility, google sheet. 

It would be worth your time to check it out!

Trust me, I'm a doctor. 

I don't believe you, don't you dare come close to my citadel... 

Share this post


Link to post
Share on other sites
591
[SWOB]
Members
2,707 posts
15,956 battles

Impressive stuff!

Only problem, and I know this is a limitation of the API....

Someone like me, with nearly 1000 games in the Gearing is forever stuck with the potato stats I had in the beginning.

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
Just now, sbcptnitro said:

Impressive stuff!

Only problem, and I know this is a limitation of the API....

Someone like me, with nearly 1000 games in the Gearing is forever stuck with the potato stats I had in the beginning.

 

That's why, (for example) my clan uses it, but they say you only have to qualify on a single ship. Think of it as proof of competency. 

 

Heck... the Gearing is my worst ship by my own creation... and I spent season two of clan battles playing exclusively the Gearing.

 

This is meant as a tool to help people, it is not meant as the gospel. It's meant to help people navigate what their stats mean.

Share this post


Link to post
Share on other sites
591
[SWOB]
Members
2,707 posts
15,956 battles
3 minutes ago, Spartias said:

That's why, (for example) my clan uses it, but they say you only have to qualify on a single ship. Think of it as proof of competency. 

Ah yes, that makes a lot of sense.

But again, impressive work! You are a googlesheet pro!

:Smile_medal:

Share this post


Link to post
Share on other sites
1,633
[-K-]
[-K-]
Members
3,223 posts
13,021 battles

I commend the time, effort, and commitment that went into producing this tool. However, I have to disagree that it is an appropriate way to gauge Clan Battle readiness.

That's because an analysis is only as good as its data. In this case, the data is from Random Battles. Particularly at high levels of competition, Clan Battle performance hinges on format-specific and teamwork-based knowledge that is simply not present in Random Battle data. In fact, a lot of performance metrics nearly invert in value for Clan Battles -- a destroyer that's getting a huge amount of torpedo hits is probably NOT providing enough screening or scouting for the CB format. Cap contests don't show up at all. Damage farmers need not apply, because it's way better if the team sinks a ship before it has time to repair.

In short, extrapolating CB readiness from RB data ignores all specialized CB-relevant knowledge, and favors certain playstyles that are not actually effective in the CB environment. It's a cool and helpful tool, but I'd caution against reading too much into it for that purpose.

  • Cool 1

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
1 minute ago, Edgecase said:

I commend the time, effort, and commitment that went into producing this tool. However, I have to disagree that it is an appropriate way to gauge Clan Battle readiness.

That's because an analysis is only as good as its data. In this case, the data is from Random Battles. Particularly at high levels of competition, Clan Battle performance hinges on format-specific and teamwork-based knowledge that is simply not present in Random Battle data. In fact, a lot of performance metrics nearly invert in value for Clan Battles -- a destroyer that's getting a huge amount of torpedo hits is probably NOT providing enough screening or scouting for the CB format. Cap contests don't show up at all. Damage farmers need not apply, because it's way better if the team sinks a ship before it has time to repair.

In short, extrapolating CB readiness from RB data ignores all specialized CB-relevant knowledge, and favors certain playstyles that are not actually effective in the CB environment. It's a cool and helpful tool, but I'd caution against reading too much into it for that purpose.

 

To an extent I agree. The following is from a duplicate thread I made on the EU. And I cautioned a lot of excitement there with the following.

 

"Clan Battles is where stats matter the least for destroyers of all things. How do you accurately quantify successfully smoking up an ally, screening them for torps, all while cap contesting and spotting... without dying? To quantify that in a manner that the average joe can agree on? 

 

Random battle stats do however work to serve as... proof of competency if it were. But where should that little grey line of competency reside? I have my opinions, and they're in the stat goals, but they're obviously subjective and should be editable by others."

Share this post


Link to post
Share on other sites
957
[TSPC]
Members
2,369 posts
7,237 battles

Playing around with this makes me realize that taking pot shots with a low chance of hitting really makes it hard to hit that category in any ship...

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
1 minute ago, enderland07 said:

Playing around with this makes me realize that taking pot shots with a low chance of hitting really makes it hard to hit that category in any ship...

Idk, something I've noticed is that almost everyone has something exceptional about their stats. Most unicums have very high survival. Most brawlers have a low survival. Main battery accuracy says more about your playstyle than it does how much you're influencing the game. 

Share this post


Link to post
Share on other sites
Members
1,592 posts
8,541 battles
1 hour ago, Edgecase said:

I commend the time, effort, and commitment that went into producing this tool. However, I have to disagree that it is an appropriate way to gauge Clan Battle readiness.

That's because an analysis is only as good as its data. In this case, the data is from Random Battles. Particularly at high levels of competition, Clan Battle performance hinges on format-specific and teamwork-based knowledge that is simply not present in Random Battle data. In fact, a lot of performance metrics nearly invert in value for Clan Battles -- a destroyer that's getting a huge amount of torpedo hits is probably NOT providing enough screening or scouting for the CB format. Cap contests don't show up at all. Damage farmers need not apply, because it's way better if the team sinks a ship before it has time to repair.

In short, extrapolating CB readiness from RB data ignores all specialized CB-relevant knowledge, and favors certain playstyles that are not actually effective in the CB environment. It's a cool and helpful tool, but I'd caution against reading too much into it for that purpose.

Not to further comment on the OP's post, but to point out this is, IMO, the issues a lot of us have with ranked, especially with DD's.

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
Just now, Kevs02Accord said:

Not to further comment on the OP's post, but to point out this is, IMO, the issues a lot of us have with ranked, especially with DD's.

 

And I agree. Though I have noticed that out of the 6 key stats that everyone wants in randoms, the ones I chose for this, dd's with high survival and high win rates generally speaking have no issues garnering the other stat categories.

 

I do feel that survival and win rate are key combinations for destroyers.

Share this post


Link to post
Share on other sites
957
[TSPC]
Members
2,369 posts
7,237 battles
48 minutes ago, Spartias said:

Idk, something I've noticed is that almost everyone has something exceptional about their stats. Most unicums have very high survival. Most brawlers have a low survival. Main battery accuracy says more about your playstyle than it does how much you're influencing the game. 

soo you're saying taking blind shots in my Montana at the beginning of games says something about my playstyle.... :)

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
7 minutes ago, enderland07 said:

soo you're saying taking blind shots in my Montana at the beginning of games says something about my playstyle.... :)

Hehe!

 

Personally I really really love first bloods. It's an immediate and delightful confidence boost to my team if the reds lose a ship first. Tactical information, such as where I'm located, is something I don't want to freely give out to the reds. It makes it far more difficult to get that critical shot in. There are so many battles where I'll hold my fire for quite a while... just to set up that first salvo. 

Share this post


Link to post
Share on other sites
1,013
[OPG]
[OPG]
Beta Testers
1,451 posts
10,574 battles
2 hours ago, Edgecase said:

I commend the time, effort, and commitment that went into producing this tool. However, I have to disagree that it is an appropriate way to gauge Clan Battle readiness.

That's because an analysis is only as good as its data. In this case, the data is from Random Battles. Particularly at high levels of competition, Clan Battle performance hinges on format-specific and teamwork-based knowledge that is simply not present in Random Battle data. In fact, a lot of performance metrics nearly invert in value for Clan Battles -- a destroyer that's getting a huge amount of torpedo hits is probably NOT providing enough screening or scouting for the CB format. Cap contests don't show up at all. Damage farmers need not apply, because it's way better if the team sinks a ship before it has time to repair.

In short, extrapolating CB readiness from RB data ignores all specialized CB-relevant knowledge, and favors certain playstyles that are not actually effective in the CB environment. It's a cool and helpful tool, but I'd caution against reading too much into it for that purpose.

I absolutely agree with you. 

100%

However...

It's much easier to take a player that shows extreme ability, proficiency, and natural talent in Randoms and then change their mindset to the proper CBs mentality than it is to take someone who is decidedly average or below average to begin with. Granted,  you will have those stubborn damage farmers who can't make the transition and they have to be culled. Still, I'd rather start with someone who can actually land their shells and mitigate damage than not. 

In that regard, this tool absolutely has utility; or any stats-measuring site/tool for that matter. 

I treat Random Battle stats like companies treat a college degree. Random Battle stats just get your foot in the door and get you an interview (like a college degree). THEN, you have to prove you can put your money where your mouth is in CBs and play as a Team for the win, sacrificing your body if need be. 

  • Cool 2

Share this post


Link to post
Share on other sites
957
[TSPC]
Members
2,369 posts
7,237 battles
Just now, Spartias said:

Hehe!

 

Personally I really really love first bloods. It's an immediate and delightful confidence boost to my team if the reds lose a ship first. Tactical information, such as where I'm located, is something I don't want to freely give out to the reds. It makes it far more difficult to get that critical shot in. There are so many battles where I'll hold my fire for quite a while... just to set up that first salvo. 

I normally try to do that behind islands. It's really fun on Two Brothers because most ships are in predictable ish areas... 

  • Cool 1

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
15 minutes ago, enderland07 said:

I normally try to do that behind islands. It's really fun on Two Brothers because most ships are in predictable ish areas... 

 

Fun is what it's all about!

Share this post


Link to post
Share on other sites
172
[WOLFB]
Members
451 posts
3,842 battles

Tim and Eric Awesome Show, Great Job! head chin facial hair forehead

This spreadsheet... blows my mind...

People make fun of me because I geek out on spreadsheets a lot.

Now I realize I am a mere child... a primate... a potted plant on the windowsill...

  • Cool 1
  • Funny 1

Share this post


Link to post
Share on other sites
3,962
[INTEL]
Members
8,815 posts
26,264 battles
On 6/10/2018 at 6:27 AM, Spartias said:

I've seen the question "What is the Greatest Stat?" asked quite a lot recently. I have my own opinions about that, and I've even voiced those opinions. (Hint: it's overall win rate with reasonable asterisks attached to the theory.)

 

But my opinions on the greatest stat isn't really what this is about. This is about giving players a way to discover and manipulate data themselves. I spend a lot of time delving into the API that Wargaming so kindly provides to us, and I like to think that gives me a certain perspective on things. This spreadsheet is designed to give that ability to you as well. With the upcoming third season of Clan Battles on our doorstep, I was hoping that this could help players understand a little better where they need to prepare themselves.

 

The first thing one does is turn the sheet on, which is done in cell B1 by entering any clan tag. It doesn't matter if you're in that clan, as entering a clan tag simply turns the sheet on. You'll then be able to either type a name into cell C5 (make sure you spell it right) or select one of the members of that clan from the drop down menu there. You don't have to search for a player in that clan, but it does make it easier.

 

SPQ-Name.JPG.c4c5aa014a02d6663e6e1f902d50ead0.JPG

 

From there the sheet will begin loading data. What you're able to manipulate is in the middle of the screen labeled "Stat Goals."

 

I think I missed something here. "loading data..." What data, from where? Does it somehow connect to WOWs???

Share this post


Link to post
Share on other sites
393
[5D]
Members
1,287 posts
7,597 battles
1 hour ago, Taichunger said:

I think I missed something here. "loading data..." What data, from where? Does it somehow connect to WOWs???

 

Yes, the spreadsheet pulls in new and up to date data from Wargaming upon each new search. The sheet uses the same API that any of the usual third party websites uses.

 

This is where the sheet gets its data.

  • Cool 1

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×