Time per board & barometer scoring

[TimG]

Two topics, one post.

 

1) I've played in two 12 board ACBL tournaments on BBO. Both times, I found them very slow -- I think the always had at least 3 minutes left at the end of the round, and usually 5-8. Is this common and if so, wouldn't it be good to shorten the rounds to 12 or 13 minutes?

 

2) Why isn't there barometer scoring? At least in a few of the games. Or, have I just missed those games?

 

OK, another one:

 

3) It appears that scoring is done across all sections. Why aren't section awards (there never seem to be overall awards) also scored across sections?

 

Tim

[Rado]

Hi Tim and welcome to BBO forums!

 

Some history hints:

- there was a long discussion at the beginning of tourney feature at BBO. There were unlimited time, 8 minutes per board (16 for 2, 27 for 3), "speedball" tourneys (5 minutes per board). Passing more than year from the first BBO tourneys it appears that the "live" formula 15 minutes per 2 boards round is the best fitted.

- many of the BBO tourneys (of which ACBL are small part) have barometer scoring

- section awards? in WBF championships this feature missing usually, in my view no need to implement at BBO:-))))))))))

Regards

Rado

[TimG]

- section awards? in WBF championships this feature missing usually, in my view no need to implement at BBO:-))))))))))

One or both of the ACBL tournaments I have entered were big enough that there were two sections. The boards were MPed across teh field, but the final results (masterpoitn awards) were done within a section. So, there were four pairs who got a section 1st, four who got a section 2nd and so on. I think we agree: there is no need for sections. MP across the field and rank the pairs across the field. (Maybe rank across a direction so that there are two winners, but the movements should be able to be designed with direction switching so that even this is not necessary.)

 

Tim

[hrothgar]

- section awards? in WBF championships this feature missing usually, in my view no need to implement at BBO:-))))))))))

One or both of the ACBL tournaments I have entered were big enough that there were two sections. The boards were MPed across teh field, but the final results (masterpoitn awards) were done within a section. So, there were four pairs who got a section 1st, four who got a section 2nd and so on. I think we agree: there is no need for sections. MP across the field and rank the pairs across the field. (Maybe rank across a direction so that there are two winners, but the movements should be able to be designed with direction switching so that even this is not necessary.)

 

Tim

Hi Tim

 

I've actually had the opposite reaction to the formats at BBO.

My major complaint is that the sections are much too large, dramatically increasing the "luck" factor within events.

 

Assume for the moment that we are looking at commercial tournaments that charge a participation fee. [i argue that this is significant because I think that there is a relation between free tournaments and players quiting mid-event]

 

I would strong prefer to have commercial tournaments automatically segment into an appropriate combination of Mitchell and Howells segments. While you can't corss compare between sections, at least you lower the crapshoot element intrinsic to a single large pool...

[TimG]

I would strong prefer to have commercial tournaments automatically segment into an appropriate combination of Mitchell and Howells segments. While you can't corss compare between sections, at least you lower the crapshoot element intrinsic to a single large pool...

So, you'd like to have lots of little sections scored independently?

 

That is perhaps a better solution, though I imagine it will transfer much of teh randomness to which section you are placed in.

 

Really, though, talk of randomness is pretty random when you're talking about a 12 board tournament.

 

Tim

[inquiry]

Really, though, talk of randomness is pretty random when you're talking about a 12 board tournament.

Right you are!!!! We need real world 26 board tourneys, expecially on weekends. I am certain i am losing because the events are long enough. :D

 

But seriously, more 16 and 20 boards (or longer) events would be good.

 

ben

[Frosty]

Hi Tim -

 

The time needed for tourney rounds tends to be relative to the connections. You might be unaware that - even when a tournament is widely publicized to have less time per round - many players will sign up who cannot play to that time constraint, which can slow down the whole event. Many times this is because connections are slow - sometimes it is because the players are slow themselves. I share your preference to avoid the long waits at the end of rounds and recommend that you look for tournaments that have shorter times per Board - 7 minutes or less. This is usually part of the tournament description.

 

Re the sectional scoring, particularly in the ACBL games - I would vote to eliminate sectional scoring and give awards based on over-all scoring. I think that the sections are set to 10 pair per direction and it is a daily occurrence that pairs win more MP's with lower over-all scores simply because they were in an easier section and fared well in their own section. If all are playing the same boards, the overall scoring by direction makes more sense, no matter how many sections you're using to run the event.

 

Frosty :)

[hrothgar]

Right you are!!!! We need real world 26 board tourneys, expecially on weekends. I am certain i am losing because the events are long enough. :)

 

But seriously, more 16 and 20 boards (or longer) events would be good.

 

ben

Ben and Tim are raising a very serious issue:

 

There is a mathematical relationship between

 

1. The total number of pairs that have entered a tournament

2. The total number of boards required to determine the best "X" pairs with confidence "Y"

 

There are a couple of ways to finesse this issue.

 

Option 1: If you have 400 pairs and you can only run a 12 board tournament, then simply simultaneously run 20 tournaments of 20 pairs each. Don't bother comparing results across sections because the information is pretty much meaningless.

 

Option 2: Alternatively, Directors have every right to create a single large tournament, however, both players and directors need to accept the tournament results will by guided as much by luck as by skill.

 

Personally, I have a strong preference for Option 1.

Your milage may vary...

[uday]

We have to choose between overall awards and section awards.

 

Unfortunately, the formula we have to use for computing awards punishes the players ( by dramatically reducing total awards) if we use overall awards.

 

We are only allowed to score our Ts as club games.

 

So, what i do is issue separate awards for NS/EW in each section, and , where feasible, i break the field up into sections

 

Clocked pairs: broken into sections of no more than 15 (16?) tables

Individuals : scored as one large section, with overalls ( here, i need a larger field so that the pairings (random) are less likely to cause playbacks

Swiss: scored as one large section ( here, NS/EW slide around, and I want a large field ).

 

I think the acbl MP awards could stand a little rework for an online environment. Any serious changes, like allowing us to run sectionals, or change the way we compute points would probably require the acbl Board to grant permission.

[xx1943]

I think the always had at least 3 minutes left at the end of the round, and usually 5-8.  Is this common and if so, wouldn't it be good to shorten the rounds to 12 or 13 minutes?

 

2) Why isn't there barometer scoring?  At least in a few of the games.  Or, have I just missed those games?

 

Tim

Hi Tim

 

if u like to play 6 min per board, there is every day ** Very Quick and Nice ** at 6:20 CET = 0:20 am EDT.

 

Swiss; 4 rounds 3 boards

 

We would appreciate to welcome you

 

Cheers

 

Al

[jwchiu]

Hmm, this has been an interesting question, ever since I saw the rare bird of a fourteen table round robin (27 rounds) :D

 

To maximize the number of masterpoints paid, while considering the number of pairs who receive points and the quality of the comparisons, alternate between standard Mitchells and Howell movements. Generally, for the sake of comparisons, the ideal movement is a seven-table Howell with 12x1 board structure.

 

Pairs Movement

12 Six table Howell movement, either 12x1 revenge or 6x2 boards

14/16 N table Howell movement, 12x1 curtailing

18/22 N table Mitchell movement, 6x2 boards

24/26 N table Mitchell movement, 12x1 boards

Alternately, split into two Howells, depending on adversity to revenge

28/32 Split into two Howells

34 Tricky, but two Mitchells works fine

36/40 Split evenly into two Mitchells

Alternately, split into three Howells, depending on adversity to revenge

42/48 Split into three Howells

50/54 Tricky, but three Mitchells work fine; alternately split into four Howells

... etc

 

We observe that the same number of points is paid out using a nine table Howell movement or a nine table Mitchell movement, due to the 1.2 masterpoint cap. This is used to provide the building blocks of small sections, used to partition the tables above.

 

**************************************************************

 

Moving on to another topic, such small sections imply the need for a general sense of seeding. Have there been attempts to seed fields? In my experience, the first pair to buy the entry is 1 N/S, the second 1 E/W, etc. If the sections were balanced, first by extremely strong players, then by average performance, then such a split would be more feasible.

 

Jason