OT2017 Round 5

Is there something different than what is on the Vassal page for C&C Ancients Giulio? I'm on Vassal 3.2.17, C&C module 3.3, and all the extensions listed below that module on that page. Let me know. Thanks.

No Dan. That's the correct reference. But when I load your log file I get the list of warnings below, suggesting you are using an old version of the extensions. I wrote you because I observed a few strange behavior in your log file and I thought that maybe the old extensions were the cause.

- Game saved with version 1.0 of extension 'CCA_C3iScenarios', you are running version 3.3. Please upgrade to the latest version of this extension.
- Game saved with version 2.2 of extension 'CCA_Expansion2', you are running version 3.3. Please upgrade to the latest version of this extension.
- Game saved with version 2.0 of extension 'CCA_Expansion3', you are running version 3.3. Please upgrade to the latest version of this extension.
- Game saved with version 1.2 of extension 'CCA_Expansion4', you are running version 3.3. Please upgrade to the latest version of this extension.
- Game saved with version 1.0 of extension 'CCA_Expansion6', you are running version 3.3. Please upgrade to the latest version of this extension

Figured it out Giulio - I had to play that game on an old computer and had forgot that is where the log file was from. Everything is up to date on the one I'm on now. Thanks for pointing it out so I can update the other.

Game 1: Britons (craniumgroup) vs Romans (stormwalker): 6-4
Game 2: Britons (Stormwalker) vs Romans (craniumgroup): 6-4

A tie score for the match. Pretty intense battles in both of them.

Ah, that was a close call! Congrats to Tomek and Brent who went to the quarterfinals.

Mark the procedure "Where there are ties for Cumulative Victory Points, the tie is determined by Ranking based upon the strength of the opponents player" is different from what is reported in the published rules of the tournament. The new procedure seems interesting but I believe we have to stick to the procedure "After each round, positions on the Ladder will be adjusted in descending order by points, and further sorted by banners won, banners lost, and blocks lost." reported in the "OT2017 Rules and Conventions". I don't know if it does change something though.

Edit 1: I did the math and something in fact changes.

Edit 2: I have to amend what I wrote: after looking at it more carefully, I don't like the tie procedure you suggest. I prefer the official one. It is simple and it gives straight incentives: even if you are losing, try to capture all medals you can and even if you are winning try to lose the least units (which is by and large quite realistic too!) The suggested procedure is instead rather brainy, opaque and not conducive to any operationalizable strategy, as it is decided "ex-post" based on what the others do. I can see its merits in a handicap based system, but not here.

Unfortunately this is not a new ranking system.......it's the one we've been using for all preliminary rounds of this Open tournament and the Open tournaments for the past three years. Swiss is actually a very well established tournament system, used originally for Chess tournaments but now in wide use for a variety of tournament formats where numbers are too large to do a round robin. CCG players may be familiar with it, as it's used for Magic tournaments. The strength of opposition calculations give a statistical approximation of a player's position against all other players in the tournament for pairing from round to round and tie breaking, as if they had done a full round robin. It's not perfect and the margins are often fine, but it's widely used and accepted.

I'm sorry there's been confusion as I didn't publish the tournament rules.....this is how the min-tournaments are run and it's probably an unfortunate cut and paste. There's actually merit in using banner and block counts to tie break as you suggest, Giulio, but it's not how this or other Open tournaments have been ranked for the past several years, and to revert to another method now would actually be breaking the integrity of the system as it's been used in this tournament so far. Maybe the tournament rules can be reviewed and if it's what people want, a different system could be used next year.

Having said all that, i feel awkward being the number cruncher and being in a position where it has an important impact on my ranking, as it does here, and particularly as there's a discrepancy between how the tournament has been run and the published rules. Michal's a great player and very deserving of progressing, and if he's keen to keep playing, I'd be very happy to cede the spot to him.

Brent.

I'm a newcomer in these tournaments. If everybody (apart me) understood that the rules adopted were the ones you now clarified I would suggest to stay with them. There's no point in being pedantic and it might well be that some information went lost when the website was changed. But if this is not the case, I think that the best solution is simply to apply the rules as written. After all, this is what people SHOULD have expected. I think everybody agrees that the fact that a different tie breaking rule was applied in the first four rounds can be considered basically ineffectual.

Hmm.. you are correct Guilio, that is the rules. This is the first time I can recall having to break a tie to get the top 8, so I have no experience of implementing these. By recall I took these from the older Open Tournament rules on the Yahoo Group website.

I have 2 issues with using the banners scored tie break;
1. The player that plays a tougher set of opponents will tend to score fewer banners.
2. Each round has a different number of banners, Round 1 was 10, round 4 was 14, and rounds 2,3 &5 were 12.

Banners won R1 R2 R3 R4 R5 TOTAL
Tomek 8 12 12 8 8 48
BrentS 7 12 12 9 11 51
mk20336 9 5 10 14 12 50

so the result of banners won tends towards the inverse of opponent strength,

I'm happy to take advice on this if anyone has any.

Dear Colleagues,

when I started OT2017 I was assuming tie-breaking procedure dependent on strength of opponents you played with. Just like it was in 2016 and 2015. I was aware this is different from Mini-Tournament and accepted this.

That said, as per rules I was aware of, I am 9th and Brent & Tomek advances to quarter-final. Full stop. I hope that will close discussion as who advances to quarter-final in OT2017.

However, the other discussion started by Giulio - what tie-breaking procedure should be - is a different topic and I think we all can share our POV. From my perspective, block count actually is not perfect here as you play with completely different opponents and actually can win more banners with less experienced ones (comparison by Mark above is a good example). Approximation which is given by "relative strength of opponents" is I think quite satisfactory, because it shows that although tied players reach the same level, the path for some of them was more difficult and that ones should be rewarded with won tie-break.