[1eyedjack]

I spend a fair amount of time helping beginners along on BBO. (I know I am not the best at it but then I don't charge a fee.)

During this time I have received an impression that is steadily reinforced with time: There is I think a growing generation of players new to the game whose only experience is playing online, generally on BBO. (This is great, BTW. Perhaps they would never have got into the game were it not for the online facility.)

One feature that many of these players have in common is an almost complete lack of understanding of the scoring rules. OK, everyone starts out that way, but it is my observation that while they would certainly like to understand the scoring, these players are still quite happy to muddle along, sometimes for years, without buckling down and getting to grips with the issue.

They absorb by osmosis some general principles:
...Making a contract is good, going down is bad, (unless it is opps in the contract) and the more so the greater the impact.
...Doubling increases the stakes (if left in), but perhaps not by a factor of 2.
...It is good to bid a game/slam if it makes. Slams are easy to understand: bid 6 or 7. Games are a bit more complex, being 3/4/5N, 4/5M, 5m. Often unclear of the logic that gives rise to that.
...Don't double them in partscores higher than 2D unless you are sure it is going down (something ethereal to do with "doubling them into game").

Using these and other maxims they "get by" tolerably well but never really understand why, with a deeper understanding being something that they feel they must brush up on some rainy day (which never seems to arrive). They may have higher priorities to understand in the game, but there comes a time when they should grasp the mettle.

It is my suspicion that the reasons for this ambivalent attitude are two-fold:
1) The computer does the scoring for you, so why bother yourself?
2) These players have never encountered rubber bridge, on which many of the duplicate scoring rules are founded.

I have made a small attempt at addressing this. I must apologise for not searching through the learning material made available on BBO which may well duplicate this attempt and if so probably far better than my version. I had a quick browse through the "learn to play bridge" series in BBO library (which incidentally I could only find via the Windows interface. If you can get there from the Flash interface there may be a better way of unhiding it), but I could not see anything comparable in the index or appendices.

I have prepared two documents available for public download:

"Scoring.pdf" is a descriptive tutorial. It doesn't contain any summary tables the like of which you will get as a spare playing card in a deck. You all have plenty of those, but just don't understand how to apply them. This is what Scoring.pdf is intended to explain, and it is available here:

http://kvisit.com/So6Qj

"IMP.XLS" is a spreadsheet that runs under MicroSoft Excel. Should work under any version from 1997 onward, possibly earlier versions. I have not tried it in Open Office, but it doesn't use any wierd stuff so may well run. What this does is allows you to enter the contract, conditions (ie vulnerability) and result, and it will calculate the aggregate score. Furthermore it will break down that aggregate total into its component parts (overtricks, tricks bid and made, undertricks, game/slam/doubled bonus). It also allows you to enter a result at a comparison table and then converts the aggregate difference into IMPs (the IMP scale is provided).

Finally, the spreadsheet will allow you to do some very basic what-if statistics, and this is illustrated by pre-completed data to illustrate another basic maxim to apply at IMPs: It is usually better to take a certain penalty in preference to a speculative game bonus, where the results are likely to be comparable, even if the game bonus may score slightly more.

Take this problem: In an IMP scored game the opponents have just bid 4D against you, non-vulnerable, which you estimate will go three down should you choose to defend (obviously doubled). You have an alternative of bidding 4H which you expect to make (vulnerable) but you are worried that it may fail. What level of probability of success justifies going for the higher scoring 4H? The data pre-entered in this spreadsheet demonstrates that you need to be about 80% sure of 4H before it becomes worthwhile bidding it, and that if you were only 50% sure of the contract then in the long term you would expect to lose about 4 IMPs per board by bidding rather than defending. The calculation is simplistic in that it assumes absolute certainty that the opponents are going exactly 3 off, that you are either making or going 1 off in 4H, and that the opponents are not going to double you, etc, etc. But it gives a slightly better result than pure gut feel, and helps you to understand how the maxim arises. (Incidentally, the corresponding analysis for an MP-scored game is NOT provided in this spreadsheet and would come to a dramatically different conclusion. If you read scoring.pdf it should explain why.)

The spreadsheet is available here:

http://kvisit.com/S3ZWKAQ

Anyway, feedback and suggestions welcome.

[Siegmund]

Anything that helps people understand the game better is a good thing.

I will just observe, though, that I've known lots of people who play casually face-to-face for years -- both at the bridge club and in rubber games -- who never come to anything like an understanding of scoring. They do learn what games are but could play the rest of their lives without realizing that doubled and down two < making game < doubled and down three at equal vulnerability, and never grasp how IMPs OR matchpoints work.

[1eyedjack]

Thanks for that.

I should add, that feedback from beginners would be particularly welcome. One feature of a good teacher is to be able to put yourself in the mind of the student - to visualise the problems as they perceive them. I don't claim any great skill in that regard. Those of us who have been familiar with scoring for decades (myself included) may find it hard to grasp which bits the beginner finds hard to master. If these documents do more to confuse than enlighten then sadly it is not a good job, however well intentioned.

Feedback from experts is also particularly welcome in order to correct any inaccuracies!

[Tola18]

There is an additional difficulty with the BBO. The very uneven results, disturbing the usual tabels.
Mostly at IMP pairs, but also Matchpoint.

It was pointed to me once by one desillusioned BBOer. The complaining pointer did exaggerated, but I think there is surely something in it...



True, in a very uneven matchpoint field you will also get many different results.

But still the tabels you are learned in courses and books, apply only rather vaguely on the BBO...

Also, some points you see after there are played 0-3 tables to compare with. Others have 15 like in a real turney.
So many results ARE skewed.

[NickRW]

Re uneven or unexpected results - it is a black art - even for experts - trying to guess what will happen at other tables. I wouldn't trouble a beginner, or even an intermediate with it too much.

I remember one hand a few months back playing a MP game with an unfamiliar partner. The auction went 1N-3N all pass. Dummy goes down and I see immediately that I have 9 tricks cold. But dummy has ♠J6432 and I have ♠KQT7. Surely, I say to myself, most people have used a transfer and opener has corrected to 4♠ - or at least responder treated the junk spade suit as a 4 carder and used stayman - with the same general effect. I see that the ♠ game is cold and probably makes 11 tricks. I say - well - I'd better play this like a maniac risking defeat for 11 tricks in NT. As luck would have it I duly make 11 tricks for a top. But then the traveller is opened and only 2 tables are in 4♠ and one or two idiots are not even in game! Had I known, I'd not have taken such a risky line. And this happened at my local club where I play twice a week - and you'd have thought I would have had a better idea of what the field would do under those circumstances. You live and learn - or not sometimes!

Nick

[1eyedjack]

Re. uneven results. I see the complaint a lot but I remain unconvinced.
It is true that if you bid and make a routine, solid game contract you score plus 2.5 IMPs or so, typically, because two other idiots have stopped in partscore and someone else has bid slam going down, when you would normally have expected to break even with zero. But the fact remains that had you NOT bid game then your IMP score contrasted with a 2.5 IMP datum would be roughly the same discrepancy as had the room bid sensibly and BOTH of your potential scores had been contrasted with a datum of zero. In the long term, for every +2.5 you get on these hands there will be a -2.5 when the oppo bids game.

[blackshoe]

I opened the spreadsheet with NeoOffice, which is basically OpenOffice with a Mac front end. No problems.

[mycroft]

As part of my "strategy for novices" thing I wrote up many many years ago (available archved (PDF)) I wrote:

Quote

Since it's hard to work all this out at the table in the five seconds you have to decide whether to sacrifice, here's a good subset to remember: 200 beats any partscore, games are 400(420) or 600(620), slams are 920/980/990 nv and 1370/1430/1440 v.  Remember doubled downtrics: 1/3/5/8/11/14 or 2/5/8/11/14

Yes, the more detailed "how this works" is nice - I *definitely* like the way it's written up, thanks! But I still think that the above will give people 90% of the scoring "decision making" data they need.

[awm]

A good summary of different forms of scoring. There are a few interesting statements regarding matchpoint scoring that may be worth mentioning.

scoring.pdf said:

In some movements some of the hands
are played a different number of times than others, and in that case you should not
convert individual hands to percentages as to do so would suggest that each hand
carries equal weight, while in fact a "top" (ie winning all the Match Points on the
hand) should carry more weight on hands that are played more times.

Certainly in the ACBL the standard procedure is to normalize such that all boards count the same. This is difficult to do manually, but easy on computer. Most often this sort of thing happens due to a board not being played at some table (No Play) or due to a movement with a large number of tables including a sitout (some boards pass through the sitout table and some do not). The idea is that pairs should not generally be penalized for having played boards which were played fewer times.

scoring.pdf said:

You get 2 Match Points for every pair that you beat, 1 Match Point for every pair whose score is identical with yours, and 0 Match Points for every pair who score better than you (in your direction).

This is true in most of the world, but not in North America. Here we award 1 match point for every pair you beat and 0.5 match points for pairs tied. The net effect is the same (just divide by two). Our way perhaps makes more logical sense (number of MP = number of pairs you beat) but would be more work to compute by hand because of the fractions.

scoring.pdf said:

MP events achieved popularity mainly because it takes a lot less effort to score up an event under the MP method if having to do it manually, without the benefit of a computer, than alternatives.

While there's some truth to this, another advantage to the MP format is that every board counts the same. At IMP scoring there are a small number of "interesting" boards where there are potential big swings (a tough to make game, a tough to bid slam, etc) whereas there are many boards which are relatively "boring" (an easy-to-bid-and-make game where only overtricks are at question). This tends to create more luck in the IMP result, especially at pairs (i.e. you want to play the strong pairs on the "boring" hands and the weak pairs on the "interesting" hands). One mistake (or one good decision by your opponents) can have a huge impact on your score at IMP pairs, whereas at matchpoints it's just one bad board and you have many chances to make it up. In fact many tournaments gave up having BaM events specifically because it was not random enough (i.e. the same teams always won).

[jonottawa]

I've often noticed that bridge players tend to be deficient at scoring.

[1eyedjack]

I have updated the "scoring.pdf" document to include a tabular presentation and to take on board some other comments from within this thread. The original link stands, just takes you to the updated doc.