WOTM 05 Results & Congratulations

[DynamicSpirit]

Sorry for the thread hijack, but while we're on the subject there's something else I don't understand about the global ranking parings. I thought the point of the parings was to not disadvantage people who can only play one game a month, which seems to imply parings should occur within a month. But that doesn't seem to have happened. For example, September's games were WOTM01 and GOTM10, so I'd have expected those games to be pared. They're not - WOTM01 (mid Sept) is pared with October's GOTM. That mismatch seems to go right through the results.

(EDIT: I posted that before AlanH reordered the parings. The months match up OK now so you can ignore my query )

 

[DynamicSpirit]

Getting an unfair advantage by actually looking at the current table, eh!

Yes. The results appeared out of order due to recent erratic finish dates. So I've changed the sort order to use start date. Now it conforms with my description

Thanks! That all makes sense now!

I'll pass on any thanks for the fact that the reordering has caused me to drop two places from 34 to 36. Perhaps next time I'll keep my gob shut...

 

[AlanH]

Pairing happens as I described, not by calendar month. The most recent game is paired with the previous game if they are different versions (vanilla/warlords), then the next two are paired and so on.

When GOTM 15 is published it will be paired with WOTM 5, then GOTM 14 with WOTM 4 etc down to GOTM 11 and WOTM 1.

When WOTM 6 is published if will be paired with GOTM 15 and the current pairings will be used again further down the table.

Pairing breaks when there are not two consecutive games for different versions, and that was why the table was confused when the sort order was by finish dates.

[cross-posted, but maybe the description will help others who are as confused as me ]

I'll pass on any thanks for the fact that the reordering has caused me to drop two places from 34 to 36.

You're more than welcome

 

[DynamicSpirit]

I think the way you've designed the paring makes for a very good system btw. And I love the UI - especially the way the scores link to the savegames but the popup hint boxes tell you about the victory condition: Very nice touch!

 

[da_Vinci]

An interesting question about pairings will be whether pairing the deity GOTM with less difficult WOTMs gives some advantage to those playing both. I think we had at least one victory in the deity game (curve-louser! ), but even with that, the relative nature of the point awards may limit any advantage to playing and using the easier game of the pair.

If it does, then the alternating pairings is not a problem. But if being able to play an easy warlords game when there is a hard vanilla game is a big advantage, we will need to think about whether matching difficulty for the pairs is a good idea, with a fixed pairing system The drawback to doing this is that more novice players would not have a less difficult game to chose in the double hard months.

dV

 

[AlanH]

Game difficulty affects the *distribution of rankings*, I guess, but the *range* is always 0 to 100, as rankings in each game are normalised to the best score/date as 100.

 

[Conquistador 63]

True if we select "Score" or even "Combined" ranking methods. However, for the "Speed" ranking method, even the highest scoring loss will score zero in the cumulative global rankings. That is surely an incentive for playing the easier games if one chooses to be more competitive in the speed rankings.

 

[Doc TK]

I never realized that the pairings changed. I should pay closer attention.

So, I should make sure any G/WOTM I blow will get paired with the one before and after alternatively forever more - thus, better get good scores on the one before and after.

BTW - the concept of pairing is great! It makes you feel like a really bad result on a game (like deity level game) isn't as bad on you.

 

[AlanH]

You are describing a change in distribution. The distribution of speed rankings has a group at rank = zero containing all lost games, and that peak is higher when there are more losses.

My statement remains true.

 

[AlanH]

So, I should make sure any G/WOTM I blow will get paired with the one before and after alternatively forever more - thus, better get good scores on the one before and after.

Well, not quite forever more. After nine months the effect of any game disappears off the rankings radar screen.

 

[DynamicSpirit]

So, I should make sure any G/WOTM I blow will get paired with the one before and after alternatively forever more - thus, better get good scores on the one before and after.

Wot? You mean like - any G/WOTM that you only get bronze instead of silver?

 

[Doc TK]

Wot? You mean like - any G/WOTM that you only get bronze instead of silver?

I was thinking of a particular deity level one.