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.

 Daedalus81 wrote:


Yea Asuryani is going to Ynnari / DE / CW / Harlies. It's a category that is probably under performing at the moment until someone finds just the right combination of soup.

Assuming there is one. The Ynarri rules may nerf the Harlequins too much, or Craftworlds could be too big of a drag. 71/33/32 vs Custodes. 71/22/43 vs Tau.

You still need to fix your methodology, Daed - when discussing WinRate%, a draw shouldn't be considered half a win.

WR% of 50% from 3 games or 75% from 6 games, for example, prove there is a problem with the method.

Aside from that, and I appreciate the small sample size, but I wouldn't've picked Daemons as causing so much trouble for Harliequins.

 Dysartes wrote:
You still need to fix your methodology, Daed - when discussing WinRate%, a draw shouldn't be considered half a win.

WR% of 50% from 3 games or 75% from 6 games, for example, prove there is a problem with the method.

Aside from that, and I appreciate the small sample size, but I wouldn't've picked Daemons as causing so much trouble for Harliequins.

What would a draw be then? Surely I should not dump out draws?

 Daedalus81 wrote:

 Dysartes wrote:
You still need to fix your methodology, Daed - when discussing WinRate%, a draw shouldn't be considered half a win.

WR% of 50% from 3 games or 75% from 6 games, for example, prove there is a problem with the method.

Aside from that, and I appreciate the small sample size, but I wouldn't've picked Daemons as causing so much trouble for Harliequins.

What would a draw be then? Surely I should not dump out draws?

Probably, you can keep them just in case there are a lot of draws percentage between certain armies.

 Daedalus81 wrote:

 Dysartes wrote:
You still need to fix your methodology, Daed - when discussing WinRate%, a draw shouldn't be considered half a win.

WR% of 50% from 3 games or 75% from 6 games, for example, prove there is a problem with the method.

Aside from that, and I appreciate the small sample size, but I wouldn't've picked Daemons as causing so much trouble for Harliequins.

What would a draw be then? Surely I should not dump out draws?

By definition, a draw is not a win. IIRC, your dataset has wins as 2, draws (generally) as 1, and losses as 0?

Your Win Rate% should be the number results of a 2 divided by number of games - that's a true value.

Depending on how many draws there are, you could consider a Draw Rate% column as well, by capturing the number of results of a 1. Assuming a draw is strictly a VP tie, I can't imagine there are huge numbers of these.

I seem to remember a couple of oddball results of 0-0 the last time this came up - not sure whether that's officially a draw or a double loss, but they'd end up as your only edge cases.

I guess what's the philosophical premise behind this? In my head a draw is representative of the competitive value of that game. Making that draw a loss removes that value and offers a skewed perspective wouldn't it?

In a horrible example - if you won one game and tied two your win rate would be 33% even if your two ties were 90-90.

I'm not disagreeing and I can change the formula, but I'm just curious as these seems a little too binary. Draws are fairly uncommon so it doesn't cause too much havoc overall, I suppose.

I think counting a draw as 0.5/50% is the right thing.

What we're looking at the win/loss ratio to tell us is how relatively powerful armies are against each other.
A "perfect" win/loss ratio for the game is exactly 50% all round.
A draw tells us in that match the two armies were perfectly even and neither could get an advantage (such as it can given the swings of individual games).
This is every bit "on the ideal" as win-one-lose-one would be.
Hence setting a draw as 50% accurately depicts what it tells us.

I agree that a draw should be worth half of a win. It's the only sensible way to track things and removing draws only creates more work for no gain in clarity.

 Daedalus81 wrote:
I guess what's the philosophical premise behind this? In my head a draw is representative of the competitive value of that game. Making that draw a loss removes that value and offers a skewed perspective wouldn't it?

But this data isn't making a prediction of what will happen, it is what is reporting what has happened. If I play three games, I can't win 50% of them - I can only win 3, 2, 1, or 0.

1.5 is right out.

If you're reporting on the actual win rate across these events - as your post claimed to do - then it needs to report the actual wins. After all, trying to say that Harlies won 75% of their 6 games against Orks makes absolutely no sense. On a 2/1/0 point split for W/D/L, they could've won 4 & drawn 1, or won 3 & drawn 3 to get that result - and those two patterns tell quite different stories about their performances.

 Daedalus81 wrote:
In a horrible example - if you won one game and tied two your win rate would be 33% even if your two ties were 90-90.

Yes, that's exactly what my win rate would be. I'd have a 1-0-2 (or 1-2-0, depending on W-L-D or W-D-L) record. In your theoretical example where a D is worth half a win, you're claiming my 1-0-2 is as good as a 2-1-0, despite my inability to close out two of the games.

Let's take your example a step further - I draw all three games 99-99, and therefore win none of them. Your current system reports that I won 50% of my games, when I failed to win any of them.

(Yes, I fully accept that managing three draws across three rounds of a tournament is supremely unlikely, but hopefully you get the point).

 Daedalus81 wrote:
I'm not disagreeing and I can change the formula, but I'm just curious as these seems a little too binary. Draws are fairly uncommon so it doesn't cause too much havoc overall, I suppose.

As I note in the initial part of the post, draws should be highlighted as their own thing when they occur. That could be adding a draw % column, or adding a column to confirm the W/L/D (or W/D/L) record. In the case of Harlies vs. Orks, a 4-1-1 record and a 3-0-3 record tell us different things that we can't see if you roll them all into a 75% "win" rate.

 Dysartes wrote:

If you're reporting on the actual win rate across these events

Statistics are nothing without a purpose or context.

Daed is not claiming to be reporting how many wins a certain faction got.
He is claiming to be reporting how successful a certain faction got.
That's close enough to 'wins' as to be almost synonymous, but they are not quite the same. Reporting draws simply as losses (ie not-wins) would be inaccurate for portraying what he intends to portray.

...a table showing the number of games played against a given opponent across n events, with a column showing their WR% (also known as a Win Rate %) isn't reporting how many wins a faction got in that pairiing?

I'd enter those gymnastics in the next World Championships, if I were you.

Daed is there a quick way to just say how many Harlequin draws there were in your dataset?

Equally though I'm not sure it matters providing the method is consistent across factions.

I can make everyone happy! Harlequins:

Daed - Much better ( ), but I think we can go even farther.

Do you have the VPs scored in all of those games for both sides?

I'm thinking it's not just the win rate, as a win by 90-to-89 is very different than wins of 90-0.

brainpsyk wrote:
Daed - Much better ( ), but I think we can go even farther.

Do you have the VPs scored in all of those games for both sides?

I'm thinking it's not just the win rate, as a win by 90-to-89 is very different than wins of 90-0.

I sure do!

Is this what you think it should look like -- average differential for wins and losses? I have to build some logic to handle tournaments who don't do 0-100 scoring as will be apparent in one of the figures.

brainpsyk wrote:
Daed - Much better ( ), but I think we can go even farther.

Do you have the VPs scored in all of those games for both sides?

I'm thinking it's not just the win rate, as a win by 90-to-89 is very different than wins of 90-0.

He probably get the data from BCP and BCP is a mess. My list for instance was mixed CWE/HARL, but my organizer listed it like pure harlequins. I played 2 games vs TS and i`m 50% WR vs them and the second game i won just because my opponent misplayed. I won vs totally new Admech player and vs drukhari player who was in a hurry and tried some interesting but bad strategy. My list was also skewed, because i knew i would probably not face nids or greater daemons, so it was optimized to have better matchups vs Tao/Custodes and i did not play single game vs them. Those small tournament statistics are interesting, but until big events happen we will not be sure about the actual WR.
My list was not very good, but still in all my games the opponent could have won them with better play. Compared to that custodes/tao won their games for 1/1.5 hours, they had only one long game when they play vs each other.

LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

And so it begins

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

 Sim-Life wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

I'd look at the Eldar Leaks, and the Eldar reality before I took the Nid Leaks too seriously.

Breton wrote:

 Sim-Life wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

I'd look at the Eldar Leaks, and the Eldar reality before I took the Nid Leaks too seriously.

eldar leaks weren't far off from the final release, scorpions got nerfed and thats the main difference

 VladimirHerzog wrote:

Breton wrote:

 Sim-Life wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

I'd look at the Eldar Leaks, and the Eldar reality before I took the Nid Leaks too seriously.

eldar leaks weren't far off from the final release, scorpions got nerfed and thats the main difference

And banshees, and warp spiders, and probably shroud runners, and and and... there was definitely more toning down than just Scorps. Eldar may still be crazy but I think it'll be like in 8th, where the absolute OP cheese is limited to several datasheets. But I could be wrong.

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

Could you give your insight as to why you think this? I don't think they're going to be any better than Tau or Custodes at the very least, but I haven't full processed the book.

 Daedalus81 wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

Could you give your insight as to why you think this? I don't think they're going to be any better than Tau or Custodes at the very least, but I haven't full processed the book.

I think if they keep Crusher Stampede, the monsters and especially the strats for the monsters are probably just too good/cheap/durable. I don't know that they will be head and shoulders above the rest of the S tier, but I'm worried. To me, the datasheets look really nice and then there are just certain things in the book (Leviathan transnid, some of the synaptic links and strats) that just push certain pieces over the edge, kind of like Custodes.

 VladimirHerzog wrote:

Breton wrote:

 Sim-Life wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

I'd look at the Eldar Leaks, and the Eldar reality before I took the Nid Leaks too seriously.

eldar leaks weren't far off from the final release, scorpions got nerfed and thats the main difference

Scorpions got nerfed? LOL gotta love GW

I think with Eldar you had "that's good" and then "but that's quite a lot of points".

With quite a few units in this book (unless the leaks are false etc) it just feels like "that's good, oh and quite a bit cheaper than comparable things in earlier books". Which is usually the first step on the road to power.

Yeah I think the biggest thing with the Nid codex is that we have points.
And while we saw the Eldar units being very powerful we didn't see the pricetag they came with.

I would be shocked if Nids don't turn out S tier. Maybe not Tau levels of broken, but they won't be far off.

 VladimirHerzog wrote:

Breton wrote:

 Sim-Life wrote:

 Eihnlazer wrote:
LMAO nid dex leaks just dropped and they are gonna blow everything out of the water.

The Nid codex is probably what would have brought me back to 40k but from the leaks about Hive Adaptions and once-per-game synapse powers on all synapse units its like they did their level best to double down on what I dislike about 9th.

I'd look at the Eldar Leaks, and the Eldar reality before I took the Nid Leaks too seriously.

eldar leaks weren't far off from the final release, scorpions got nerfed and thats the main difference

The picture I saw had the Avatar as a LOW.

Breton wrote:


The picture I saw had the Avatar as a LOW.

so you're saying the avatar got buffed?



Automatically Appended Next Post:

EviscerationPlague wrote:

Scorpions got nerfed? LOL gotta love GW

the 30pts exarch power that let him do mortals on 5+ used to apply to the whole squad, got changed to exarch only (and kept the ridiculous pts cost)

 VladimirHerzog wrote:

Breton wrote:


The picture I saw had the Avatar as a LOW.

so you're saying the avatar got buffed?



Automatically Appended Next Post:

EviscerationPlague wrote:

Scorpions got nerfed? LOL gotta love GW

the 30pts exarch power that let him do mortals on 5+ used to apply to the whole squad, got changed to exarch only (and kept the ridiculous pts cost)

I was wondering why this power coast 30 pts, having mortals on 5++ seem insane for me, but 30 pts for the exarch is overpriced by a mile. They could have atleast made it like 10-15 pts, so atlest someone took it.