MiBasse
Joined: Dec 04, 2004
|
 Posted:
Dec 03, 2011 - 09:17 |
 
|
| VoodooMike wrote: |
| pythrr wrote: | | most clawpomb teams (bar cds) suck at low TV (unless they minmax hard) |
As it happens, most clawpomb races, with the exception of chaos, don't get a lot better as their TVs increase. Chaos does, in a huge way. Chaos Dwarves actually lose more often as TV increases.. though "more often" doesn't mean they drop below 50% overall. Chaos Dwarves are a monster team.
|
I find this pretty interesting. I suspected it for CDs and Pact - CDs because clawpombers get better against them, Pact because they are really not that good of a team to begin with - but Nurgle surprise me.
I wonder if Nurgle would hold up better in a different environment like Ranked, or just one where the meta is geared less towards bashers.
A shame we don't have the info for ranked to compare with. |
|
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Dec 03, 2011 - 09:58 |
|
|
Wreckage
Joined: Aug 15, 2004
|
| Posted:
Dec 03, 2011 - 11:00 |
|
I'd love to get more information out of this... Like what are the teams the T1 teams performed badly against?
And what are the teams the T3 races performed good against? |
|
|
MiBasse
Joined: Dec 04, 2004
|
| Posted:
Dec 03, 2011 - 11:05 |
|
| Wreckage wrote: | I'd love to get more information out of this... Like what are the teams the T1 teams performed badly against?
And what are the teams the T3 races performed good against? |
I suspect the T1 races primarily performed poorly against each other, but yes, numbers might be interesting. |
|
|
dode74
Joined: Aug 14, 2009
|
| Posted:
Dec 03, 2011 - 11:15 |
|
| koadah wrote: | Lies & damn lies. 
The stats are what they and the OP does explain what they are.
I guess they are just not what a lot of people are interested in.
To many people on Fumbbl tier 1 would be more like how often does the race appear on lists like this one |
I suspect it's more the case that the statistics do not affirm what some FUMMBLers believe to be true.
The fact that they have nothing concrete to which we can refer to back up that belief won't dissuade the die-hards, of course, and VM providing some stats derived from real data (rather than intuition derived from experience) is merely an opportunity for people to deride methodology, the concept of statistics, or simply the fact that he hasn't played much on here. But it's ok - the world really is flat  |
|
|
Wreckage
Joined: Aug 15, 2004
|
| Posted:
Dec 03, 2011 - 11:32 |
|
| dode74 wrote: | | koadah wrote: | Lies & damn lies.
The stats are what they and the OP does explain what they are.
I guess they are just not what a lot of people are interested in.
To many people on Fumbbl tier 1 would be more like how often does the race appear on lists like this one |
I suspect it's more the case that the statistics do not affirm what some FUMMBLers believe to be true.
The fact that they have nothing concrete to which we can refer to back up that belief won't dissuade the die-hards, |
Well I really have no idea what the rest of the community believes about that. I for my part have never really thought about the issue in terms of: What race performs well against most others?
I mean I greatly appreciate these numbers and I find them very interesting.
But they are not the only data the community has at hand. The site produces plenty of data at various ranges and they have been subject of debate before.
Numbers aren't some kind of magical wand you can draw to solve random problems. IF you apply them to another science they are always subject to interpretation. One could say they can sharpen the intution, allow to watch for things that would have been missed but not replace it. It's always a mistake to blindly rely on data and to believe you don't have to think for yourself anymore. Even mathematicians have a view on things and that's what they base their calculations on.
It's as voodo says: If people say tons of stuff purely on intuition it must be aswell more then appropriate to present ideas with a numerical backing. But that doesn't turn stuff into some godly truth. |
|
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Dec 03, 2011 - 11:36 |
|
|
dode74
Joined: Aug 14, 2009
|
| Posted:
Dec 03, 2011 - 11:46 |
|
It's not flat - I've seen the curvature myself
If experts provide you with the means to check their results then it's not the experts you are trusting.
I tell my flying students the opposite. "Seat of the pants" can kill you. It can, in the right situation with the relevant experience, be a godsend, but it is certainly not the first resource for an answer.
As for numbers not being a "magic wand", that was never suggested. They are what they are, and if they run counter to intuition then there is more than likely some sort of cognitive bias colouring the intuition - revisit both sources. |
|
|
Carnis
Joined: Feb 03, 2009
|
| Posted:
Dec 03, 2011 - 12:57 |
|
| RedFish wrote: | | VoodooMike wrote: |
| pythrr wrote: | | most clawpomb teams (bar cds) suck at low TV (unless they minmax hard) |
As it happens, most clawpomb races, with the exception of chaos, don't get a lot better as their TVs increase. Chaos does, in a huge way. Chaos Dwarves actually lose more often as TV increases.. though "more often" doesn't mean they drop below 50% overall. Chaos Dwarves are a monster team.
|
I find this pretty interesting. I suspected it for CDs and Pact - CDs because clawpombers get better against them, Pact because they are really not that good of a team to begin with - but Nurgle surprise me.
I wonder if Nurgle would hold up better in a different environment like Ranked, or just one where the meta is geared less towards bashers.
A shame we don't have the info for ranked to compare with. |
Coaching preference holds back nurgle. Good coaches do great with nurgle, but most good coaches go for chaos. It's more vanilla with apothecary, higher ag on the CWs.
Nurgle draws in a lot of coaches who dont really play to win & go for 100% carnage. Even there we have huge variety, the thing is, its a lot HARDER to skill Nurgle warriors than Chaos warriors, though when its done, the Nurg warriors are in general better off.
So its not surprising at all that geared up chaos start to power up, whilst geared up nurgle may start stacking skills on the BoN & may have 1-3 skill NWs holding the team back as opposed to MA5 AG3 CWs who are always well skilled by any coach.
Also nurgle seems to draw more flavour coaches, as they are a more flavoured team.. So builds like surehands kor twoheads fa are seen and the like. |
|
|
Carnis
Joined: Feb 03, 2009
|
| Posted:
Dec 03, 2011 - 13:05 |
|
Also to comment on the lizardmen, really they are I think a tier1 race clearly. Facing any liz team with 3+ block guard mb sauri and 2+ block mb po sauri will make you wish you'd drawn a different opponent. Even if you are playing clawpomb chaos dwarves, lizzies can really block you in with their 7-8 ST4+.. |
|
|
sann0638
Joined: Aug 09, 2010
|
| Posted:
Dec 03, 2011 - 13:50 |
|
I think what people would like to see is a 24 v 24 win percentage (with draw = 0.5 win) at the different TV ranges.
Ranked would be harder as the "TV range" could be more different for the two teams. With box it's about the same (obviously) so an average TV could be used to put them into a range. |
_________________
NAF Ex-President
Founder of SAWBBL, Wiltshire's BB League on Facebook and Discord |
|
VoodooMike
Joined: Nov 07, 2010
|
| Posted:
Dec 04, 2011 - 13:00 |
|
| RedFish wrote: | | A shame we don't have the info for ranked to compare with. |
We do, in fact, have information about ranked, but there are far fewer games in ranked than in box, and that makes the results harder to work with. Even in box the unpopular races have, at many TV ranges, too few games to reliably work with.
| Wreckage wrote: | I'd love to get more information out of this... Like what are the teams the T1 teams performed badly against?
And what are the teams the T3 races performed good against? |
T3 races are easy.. they're good against each other, more or less. T1 are poor against some other T1 teams and one or two T2 teams. I'll try to figure out a good way to post the data without it being a 24-post threadwall of text.
| Wreckage wrote: | It's always a mistake to blindly rely on data and to believe you don't have to think for yourself anymore. Even mathematicians have a view on things and that's what they base their calculations on.
It's as voodo says: If people say tons of stuff purely on intuition it must be aswell more then appropriate to present ideas with a numerical backing. But that doesn't turn stuff into some godly truth. |
It's never a mistake to rely on data - it's a mistake to ignore data in favour of "feelings". If numbers disagree with your expectation you should question the expectation, not the numbers. You don't have to care about numbers - you can absolutely play everything by ear, but in the long run your intuition will fall short of real numbers.
You should read the book "Moneyball" sometime - it is about this very thing. I haven't seen the movie, so I don't know if it's worthwhile.
| koadah wrote: | | A bit of intuition and gut feel can save your life so don't take it so lightly. |
Even a broken clock is right twice a day.
| Carnis wrote: | | Coaching preference holds back nurgle. Good coaches do great with nurgle, but most good coaches go for chaos. It's more vanilla with apothecary, higher ag on the CWs. |
That kind of skew is unlikely to affect things significantly if you have a sufficiently large number of games to work with.
| sann0638 wrote: | | I think what people would like to see is a 24 v 24 win percentage (with draw = 0.5 win) at the different TV ranges. |
What actual good is that sort of table? I've seen people make them before, but they just look like a wall of numbers that don't say anything. |
|
|
VoodooMike
Joined: Nov 07, 2010
|
| Posted:
Dec 04, 2011 - 13:14 |
|
ZIP of summaries by race
That's a set of text files and jpgs for each race, showing the win percentages against each race, overall win percentage for all games, and a curve based on win percentages for each race based on the team's TV. |
|
|
Carnis
Joined: Feb 03, 2009
|
| Posted:
Dec 04, 2011 - 13:39 |
|
| VoodooMike wrote: | ZIP of summaries by race
That's a set of text files and jpgs for each race, showing the win percentages against each race, overall win percentage for all games, and a curve based on win percentages for each race based on the team's TV. |
The graphs are quite a bit more descriptive than the final numbers you've presented. For instance, the amazons are only playing TV 1000-1500 basicly, so they are sweetspotting.. Lizardmen in your graph are mostly in the 1000-1600 range as well.
Maybe you could display your average win% graphs for each race/TV-range overlayed in groups of 6? 24 would probably be a bit unreadable..
Also there are other trends beyond the win% going sky high for high tv chaos, which you already presented. Dark Elves seem to do better high tv as well.
Edit:
In your graphs
Dwarves also go steeply below 40% at the higher TV ranges, but due to sweetspotting in the final numbers they end up a tier1 race despite their obvious weakness in this ruleset. So there's obviously some need for interpretation into the numbers.. |
|
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Dec 04, 2011 - 14:19 |
|
| VoodooMike wrote: |
| koadah wrote: | | A bit of intuition and gut feel can save your life so don't take it so lightly. |
Even a broken clock is right twice a day.
|
A doctor (expert) cannot actually feel what the patient is feeling. The doctor may be seeing the person for the first time where as the patient has lived with that body all their life.
Scans don't show up everything.
The person who accepts the reassurance of the experts and their data is often diagnosed too late and dies. Whereas the person who kicks up a fuss because of their gut feel/intuition can get their diagnosis on that 3rd or 4th opinion.
Of course the government accountants will say don't run those expensive tests because the data shows that 99% of the time it's nothing. But that's no comfort to the 1%.
Question everything. |
_________________
O[L]C 2016 Swiss! - April ---- All Star Bowl - Teams of Stars - 2 more teams needed |
|
|
|
| |
Elf Draft Coach
Search
