Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

Hi I have made a graph that might be interesting for those who are into ranking armies. It is shows an alternative way to look at things.
The graph shows how much players the best ranking player* of that army left behind him. I like how it focuses on groupings of tiers rather then the top 10. Have fun with it.
Also am I the only one interested in how the sisters and Inquisition players came that high ?



[edit] Added a second graph.

The following graph is based on the 5 top ranking players per codex*
The score is calculated by: (Total number of players - Ranking player)
The codexes are sorted on the average score of the selected players

*Codex is based on the publication later shown and isn't corrected for allies etc. If the codex isn't played/listed enough the graph shows less bars.

Did they ban the skiitari pods? Why are the skits coming in so low?

 oldzoggy wrote:
Hi I have made a graph that might be interesting for those who are into ranking armies. It is shows an alternative way to look at things.
The graph shows much players the best ranking player of that army left behind him. I like how it focuses on groupings of tiers rather then the top 10. Have fun with it.
Also am I the only one interested in how the sisters and Inquisition players came that high ?

I'm sorry, but I'm confused. I'd add a title and a note to the left. What is this graph comparing?

( Ranking of the best placed player with that army) -( the amount of players in the tournament).

Not sure if this is possible but does this account for armies where the primary is far smaller than the ally/additional detachments?

I am not sure what this means...explain better please.

If I understand it correctly, its basically the highest ranking person from each faction. So the top eldar player left behind 294(3rd through 292), the top ranking daemon player left behind 291(4th through 292, I think)

Basically the message for this should be if you play one of the armies that had someone place in the top 30, you can officially blame yourself because someone else managed to do it.

It's a kind of backwards way of splicing the data but sometimes that can be useful.

Where did you get the LVO results?

 Eldarain wrote:
Not sure if this is possible but does this account for armies where the primary is far smaller than the ally/additional detachments?

I just used the armies listed of the website. I have no idea how they handle allied armies.

Automatically Appended Next Post:

 vercingatorix wrote:


Where did you get the LVO results?

here
http://bcp.modelingforadvantage.com/event/lvo2016

This was just me wanting to show the results in an other perspective. Most of the discussions focussed on the 10 seemed to miss the overview.
I make no claims on why there is such a sharp difference. It might be that the armies are plain better or it might be that competitive players who know what they are doing in this meta just like to play certain armies who knows.
This clearly shows us that the large difference between high-mid-low tier performances of the armies dwarfs those differences of the top armies.

It might be interesting to add some sort of error bars to the graph by incorporating the spread of the results per army to make my point even clearer. We might even calculate what armies did it significantly better then others. My guess is that this would explain the result of sisters and inquisition, and clearly groups tiers of armies, but that feels like work. ; )

um... couldn't you have just given the rank of the highest player in each race? That essentially what your graph tells us but is way more confusing

aka;
Eldar: 1
Space marine: 2
Daemons: 4

etc...

No not at all.
The distance between the armies is what is important not the actual place. Lets give an example why. Suppose you tried ranking these ships by size. You ranked them by using 1, 2, 3... etc. Discussions like Cruiser A is so much bigger / OP then cruiser B since it is 1st and not 3th would not make much sense since they are nearly the same size while each of them dwarfs the rest of the ships.


I agree that this graph is a bit lazy. Applying some decent statistics etc would have been much better but I don't like to do those kind of calculations in my spare time.

 oldzoggy wrote:
No not at all.
The distance between the armies is what is important not the actual place. Lets give an example why. Suppose you tried ranking these ships by size. You ranked them by using 1, 2, 3... etc. Discussions like Cruiser A is so much bigger / OP then cruiser B since it is 1st and not 3th would not make much sense since they are nearly the same size while each of them dwarfs the rest of the ships.

Spoiler:

I agree that this graph is a bit lazy. Applying some decent statistics etc would have been much better but I don't like to do those kind of calculations in my spare time.

what you said makes absolutely no sense. Its no different than if ships 1 & 2 were the same type of ship.

Saying the top ranked skitari player ranked 144 gives me the exact same information as saying it placed better than 150 other people

Saying skitari ranked 143 gives me the exact same information as saying it placed better than 144 other people

It doesn't and this is why.

The thing that is missed in this evaluation is spread of primary codexes from the front rank to last. Just because the top ranked person did well does not mean the rest of the codex did well. I think doing what is done in Cross Country races could apply here: take the top five finishers per primary codex, add up their ranks and compare from there. If a particular codex had less than 5 participants in the tournament, the remaining "players" all finish in 296th place, one above the number of participants.

With this in mind, here's the spread from top to bottom:
5th place in Top 50
Eldar: 1+2+6+8+16 = 33pts
Necrons: 5+7+8+34+38 = 92pts
Space Marines: 3+19+24+25+36 = 107pts
Dark Angels: 8+11+26+41+46 = 132pts
Tau: 12+18+26+31+46 = 133pts
Renegades: 17+28+28+30+43 = 146pts

5th place in Top 100
Daemons: 4+13+37+71+88 = 213pts
Cult Mechanicus: 13+13+40+70+81 = 217pts

5th place in Top 150
Dark Eldar: 34+57+57+76+137 = 361pts
Tyranids: 43+48+71+85+121 = 368pts

5th place in Top 200
Imperial Knights: 63+63+91+103+161 = 481pts
Orks: 54+89+116+132+146 = 537pts
CSM: 21+113+116+132+173 = 555pts
Astra Militarum: 101+106+121+146+146 = 620pts

5th place in Top 250

5th place in Top 295
Space Wolves: 80+97+116+126+251 = 670pts
Khorne Daemonkin: 99+132+137+146+188 = 702pts
Adeptas Sororitas: 32+76+137+214+278 = 737pts
Blood Angels: 93+108+153+194+196 = 744pts
Grey Knights: 21+163+216+235+241 = 876pts

Did not have 5 participants
Inquisition: 23+233+296+296+296 = 1144pts
Harlequins: 137+233+296+296+296 = 1258pts
Skitarii: 181+296+296+296+296 = 1365pts


Note that a Top 25 participant army wound up at the bottom of the Top 295 grouping because the other four participants were significantly behind in the rankings. This provides a better overview of performance and allows for some evaluation.

Some other notes include the inclusion of Orks and Astra Militarum within the "5th place in Top 200" category even though their 5th place finisher would slot them in the "5th place in Top 150" category. Because they were so bottom loaded on rankings, they actually finished further behind some armies whose 5th place finisher was over #150. I thought it best to do the groupings according to overall point total rather than 5th place finish.

For the record, there were exactly five Adeptas Sororitas participants in the tournament, something I didn't know would happen when I started. X-Country has always counted the Top 5 finishers, so that's why I chose that number.

Did you correct for the fact that the 32nd place player was not actually sisters?

 oldzoggy wrote:

Saying skitari ranked 143 gives me the exact same information as saying it placed better than 144 other people

It doesn't and this is why.

Spoiler:

In the ranking case you know that there are at least 143 players. In the reversed version you know how many players there are above it (since you know rank 1) and below it and thus know the ranking and the total amount of players


Example why this matters

Suppose there is a tournament with 4 types of armies and the player with the best ranking per army is 1, 5,7, 10 and 20. This doesn't tell you how much players where attending at the tournament. This does matter a lot.
If there where just 30 then the 4th type of army / player would bad at the game. The situation would be quite different if there where 300 players in this tournament.

ok fine lol. Then just say what I said as well as how many people were in the tournament.

 Tinkrr wrote:
Did you correct for the fact that the 32nd place player was not actually sisters?

Nope sorry. I just went by the puplication of that site. I would love to somehow correct it for that do you have more info on lists ?
I was wondering the same thing of the inq list. Usually they don't score that well.





Automatically Appended Next Post:

Tropic Thunder wrote:
The thing that is missed in this evaluation is spread of primary codexes from the front rank to last. Just because the top ranked person did well does not mean the rest of the codex did well. I think doing what is done in Cross Country races could apply here: take the top five finishers per primary codex, add up their ranks and compare from there. If a particular codex had less than 5 participants in the tournament, the remaining "players" all finish in 296th place, one above the number of participants.

With this in mind, here's the spread from top to bottom:
5th place in Top 50
Eldar: 1+2+6+8+16 = 33pts
Necrons: 5+7+8+34+38 = 92pts
Space Marines: 3+19+24+25+36 = 107pts
Dark Angels: 8+11+26+41+46 = 132pts
Tau: 12+18+26+31+46 = 133pts
Renegades: 17+28+28+30+43 = 146pts

5th place in Top 100
Daemons: 4+13+37+71+88 = 213pts
Cult Mechanicus: 13+13+40+70+81 = 217pts

5th place in Top 150
Dark Eldar: 34+57+57+76+137 = 361pts
Tyranids: 43+48+71+85+121 = 368pts

5th place in Top 200
Imperial Knights: 63+63+91+103+161 = 481pts
Orks: 54+89+116+132+146 = 537pts
CSM: 21+113+116+132+173 = 555pts
Astra Militarum: 101+106+121+146+146 = 620pts

5th place in Top 250

5th place in Top 295
Space Wolves: 80+97+116+126+251 = 670pts
Khorne Daemonkin: 99+132+137+146+188 = 702pts
Adeptas Sororitas: 32+76+137+214+278 = 737pts
Blood Angels: 93+108+153+194+196 = 744pts
Grey Knights: 21+163+216+235+241 = 876pts

Did not have 5 participants
Inquisition: 23+233+296+296+296 = 1144pts
Harlequins: 137+233+296+296+296 = 1258pts
Skitarii: 181+296+296+296+296 = 1365pts


Note that a Top 25 participant army wound up at the bottom of the Top 295 grouping because the other four participants were significantly behind in the rankings. This provides a better overview of performance and allows for some evaluation.

Some other notes include the inclusion of Orks and Astra Militarum within the "5th place in Top 200" category even though their 5th place finisher would slot them in the "5th place in Top 150" category. Because they were so bottom loaded on rankings, they actually finished further behind some armies whose 5th place finisher was over #150. I thought it best to do the groupings according to overall point total rather than 5th place finish.

For the record, there were exactly five Adeptas Sororitas participants in the tournament, something I didn't know would happen when I started. X-Country has always counted the Top 5 finishers, so that's why I chose that number.

Thx I totally agree with you that this is a better approach

 oldzoggy wrote:

 Tinkrr wrote:
Did you correct for the fact that the 32nd place player was not actually sisters?

Nope sorry. I just went by the puplication of that site. I would love to somehow correct it for that do you have more info on lists ?
I was wondering the same thing of the inq list. Usually they don't score that well.

Yup, the Sisters list was mislabeled, it was actually Salamanders. The Renegades lists were also an interesting mix of stuff.

I don't know about the inquisition list as far as I know, but Grey Knights is legit, though has lots of allies.

Check out the link in my signature, it has a good chunk of the top 50 lists from the LVO. Almost all user submitted so far which is awesome and I'm always working on expanding it.

I'll just be borrowing this to justify my bitching about IG needing a new codex...

GreenShoes wrote:
I'll just be borrowing this to justify my bitching about IG needing a new codex...

^ This

Lol I wonder what Tau would look like if they weren't nerfed hahaha they still scored amazingly

Tropic Thunder wrote:
The thing that is missed in this evaluation is spread of primary codexes from the front rank to last. Just because the top ranked person did well does not mean the rest of the codex did well. I think doing what is done in Cross Country races could apply here: take the top five finishers per primary codex, add up their ranks and compare from there. If a particular codex had less than 5 participants in the tournament, the remaining "players" all finish in 296th place, one above the number of participants.

With this in mind, here's the spread from top to bottom:
5th place in Top 50
Eldar: 1+2+6+8+16 = 33pts
Necrons: 5+7+8+34+38 = 92pts
Space Marines: 3+19+24+25+36 = 107pts
Dark Angels: 8+11+26+41+46 = 132pts
Tau: 12+18+26+31+46 = 133pts
Renegades: 17+28+28+30+43 = 146pts

5th place in Top 100
Daemons: 4+13+37+71+88 = 213pts
Cult Mechanicus: 13+13+40+70+81 = 217pts

5th place in Top 150
Dark Eldar: 34+57+57+76+137 = 361pts
Tyranids: 43+48+71+85+121 = 368pts

5th place in Top 200
Imperial Knights: 63+63+91+103+161 = 481pts
Orks: 54+89+116+132+146 = 537pts
CSM: 21+113+116+132+173 = 555pts
Astra Militarum: 101+106+121+146+146 = 620pts

5th place in Top 250

5th place in Top 295
Space Wolves: 80+97+116+126+251 = 670pts
Khorne Daemonkin: 99+132+137+146+188 = 702pts
Adeptas Sororitas: 32+76+137+214+278 = 737pts
Blood Angels: 93+108+153+194+196 = 744pts
Grey Knights: 21+163+216+235+241 = 876pts

Did not have 5 participants
Inquisition: 23+233+296+296+296 = 1144pts
Harlequins: 137+233+296+296+296 = 1258pts
Skitarii: 181+296+296+296+296 = 1365pts


Note that a Top 25 participant army wound up at the bottom of the Top 295 grouping because the other four participants were significantly behind in the rankings. This provides a better overview of performance and allows for some evaluation.

Some other notes include the inclusion of Orks and Astra Militarum within the "5th place in Top 200" category even though their 5th place finisher would slot them in the "5th place in Top 150" category. Because they were so bottom loaded on rankings, they actually finished further behind some armies whose 5th place finisher was over #150. I thought it best to do the groupings according to overall point total rather than 5th place finish.

For the record, there were exactly five Adeptas Sororitas participants in the tournament, something I didn't know would happen when I started. X-Country has always counted the Top 5 finishers, so that's why I chose that number.

This provides so little meaningful data.
Because its not normalized for player count, at all.

The fact the "no 5 participant" armies got hit so hard for not having 5 participats is the most obvious example. the fact there were few main skitarii players means not that the army as a whole is bad, but only that there were little players who took it as it's main. not a hige shocker given its a relativly new, and not overly fleshed out codex. (they don't even have HQs!) Quinns are the same. Inquisition even more so given that its really more of a "small allied force" rather than a stand alone army.

On the other scale, the ones with the many participants get free "rank boost" form the mere fact all the players preforming poorly got discounted. so if you had 25 players of an army, and 5 did great while 20 were rock bottom, the army would still be ranked as if only the great ranks exist. this is probably has the greatest effect in armies with large player bases.


I'd be much more intrested to see the avarage placing of each army for example, but that still does not cover all the many odd builds where you got a tiny "main faction" and a much larger "secondary" being counted as the main, or even what people took as secondaries. (for example, skaitaii may almost get no participation, but what if it was allied for every single top-contender IoM army? this data won't be shown)


Also-why are we counting all the players who did not even have full participation? there are players who payed two, three or four games. they naturally don't provide much meaningful data.

Reavas wrote:
Lol I wonder what Tau would look like if they weren't nerfed hahaha they still scored amazingly

If by amazingly you mean the worst of the 7.5 edition codexes than yeah (Eldar, Space Marine, Necron, and DA all had 1-3 higher places for each army than tau)

notredameguy10 wrote:

Reavas wrote:
Lol I wonder what Tau would look like if they weren't nerfed hahaha they still scored amazingly

If by amazingly you mean the worst of the 7.5 edition codexes than yeah (Eldar, Space Marine, Necron, and DA all had 1-3 higher places for each army than tau)

Pretty clear that he meant that on average they did very well compared to the overall field. Their top 5 placed significantly higher on average than the majority of the other armies. Does that mean that they did the best out of all the armies? No. But there is no reason why they should have. It's not like GW is supposed to just have each new codex/update be better than all the others that preceded it in 7th edition. Tau is kicking it just fine.

I would not be that surprised that you could not find a significant difference between Space Marines, Tau, Necrons, Dark Angels and Renegades when you apply decent statistics on the data of the second graph.

 luke1705 wrote:

notredameguy10 wrote:

Reavas wrote:
Lol I wonder what Tau would look like if they weren't nerfed hahaha they still scored amazingly

If by amazingly you mean the worst of the 7.5 edition codexes than yeah (Eldar, Space Marine, Necron, and DA all had 1-3 higher places for each army than tau)

Pretty clear that he meant that on average they did very well compared to the overall field. Their top 5 placed significantly higher on average than the majority of the other armies. Does that mean that they did the best out of all the armies? No. But there is no reason why they should have. It's not like GW is supposed to just have each new codex/update be better than all the others that preceded it in 7th edition. Tau is kicking it just fine.

Yes, tau are doing "just fine"
But "just fine" is not the realm of premature nerfs and additional irrational nerfs voted further (and most of tau nerfs are not even RAW questionable, but straight faced rule changes) being justified, yet that's what's going on.

 BoomWolf wrote:


I'd be much more intrested to see the avarage placing of each army for example, but that still does not cover all the many odd builds where you got a tiny "main faction" and a much larger "secondary" being counted as the main, or even what people took as secondaries. (for example, skaitaii may almost get no participation, but what if it was allied for every single top-contender IoM army? this data won't be shown)

Also-why are we counting all the players who did not even have full participation? there are players who payed two, three or four games. they naturally don't provide much meaningful data.

Ask and you shall recieve. No, I haven't corrected for tiny main factions and allies and stuff, I just went off their results and positioning (1st, 2nd, etc, the way written on their site). Standard Error Included.



So, Renegades score highest on average, edging out Eldar even.
All factions with low representation should be treated with scepticism, due to lack of data.
Interesting is the consistency of Skitarii, not to be treated as solid, but unusual for such a small sample.

Eldar, Space Marines and Orks are the most robust codices, Orks being surprising in that regard. Obviously, Inquisition is the swingyest, unsurprising due to small sample size.

Anyway, have fun with it.

That's actually very interesting.
The fact renegades are up top does not surprise me much. It's a solid army list, and uses all the units the meta says are "useless"-hordes, artillery spam, disposable units, cheap assault units and such, meaning nobody brings the tools to handle it. Many armies just lack the number of shots/attacks necessary as all the points are spent on quality that does nothing.

Eldar superiority among the "meta" armies also is rather predictable. Admech, tau, impknights and crons being not far behind also predictable.

Dark angels are the surprising one in my eyes. "Internet wisdom" says it's inferior to codex marines, yet it preformed better.
Khorne's poor performance also unexpected. I didn't think they would rock the house, but I expected mid pack at the least.

Say farseer, do you have easily manageable spreadsheets there? Because if so I'd also be interested in seeing how it looks in "grand alliance" averages (all eldar together, all chaos together and all Imperium together.)

 BoomWolf wrote:

 luke1705 wrote:

notredameguy10 wrote:

Reavas wrote:
Lol I wonder what Tau would look like if they weren't nerfed hahaha they still scored amazingly

If by amazingly you mean the worst of the 7.5 edition codexes than yeah (Eldar, Space Marine, Necron, and DA all had 1-3 higher places for each army than tau)

Pretty clear that he meant that on average they did very well compared to the overall field. Their top 5 placed significantly higher on average than the majority of the other armies. Does that mean that they did the best out of all the armies? No. But there is no reason why they should have. It's not like GW is supposed to just have each new codex/update be better than all the others that preceded it in 7th edition. Tau is kicking it just fine.

Yes, tau are doing "just fine"
But "just fine" is not the realm of premature nerfs and additional irrational nerfs voted further (and most of tau nerfs are not even RAW questionable, but straight faced rule changes) being justified, yet that's what's going on.

When that "just fine" occurs with those rulings in effect, that's evidence that the rulings are reasonable, not the other way around.

+1 These results show that The top armies except perhaps eldar are all closely matched this is a good thing. The only thing that could make it better is when the other armies would be just as closely matched.

Armies clearly dominating is a bad thing.

 BoomWolf wrote:
That's actually very interesting.
The fact renegades are up top does not surprise me much. It's a solid army list, and uses all the units the meta says are "useless"-hordes, artillery spam, disposable units, cheap assault units and such, meaning nobody brings the tools to handle it. Many armies just lack the number of shots/attacks necessary as all the points are spent on quality that does nothing.

Eldar superiority among the "meta" armies also is rather predictable. Admech, tau, impknights and crons being not far behind also predictable.

Dark angels are the surprising one in my eyes. "Internet wisdom" says it's inferior to codex marines, yet it preformed better.
Khorne's poor performance also unexpected. I didn't think they would rock the house, but I expected mid pack at the least.

Say farseer, do you have easily manageable spreadsheets there? Because if so I'd also be interested in seeing how it looks in "grand alliance" averages (all eldar together, all chaos together and all Imperium together.)

If your list can deal with spammed Gorepacks, then Daemonkin are trash, since competitively their lists are literally just a Str.D 'Thirster, Be'lakor for Invis, min Slaughter Cult & min/maxed Gorepacks until pts run out.

They're really like Tyranids in the sense that you only need to be capable of dealing with 4-5 Flyrants.