[inquiry]

In the vote for what you would like to have discussed, a comparison between imps and matchponts now has enough votes to earn its own thead. We should discuss in this thread, the mechanics and scoring of the two games, but then get into replies that cover matchpoint strategy versus imp strategy when it comes to bidding games, bidding slams, safety plays, light overcalls, risky preempts, and the like.

I think we can count on some "heated" discussion when it comes to what it the optimal strategies on specific example hands one game versus the other. don't be confused by this... the one thing all good bridge players can agree on is that they will disagree with each other on almost everying related to the game... :-)

Just find the strategy that makes the most sense to you and see if you can apply it at the table...

Ben

[inquiry]

This is the body of a thread I wrote a long time ago...

Imps is one way to compare duplicate results.

You could play total points. You play the hand, and get say 620, at the "other" table they are down one for -100 so your ntet score (total points) is 720. But there are some really wild swings, so one bad hand (say you bid 7H and they set it with a ruff at trick one while 7NT bid at the other table, would lead to such a large total point lead that, well, you lose pretty much regardless how well you play the rest of the hands.

Rather than allow one hand or two key hands decide everything, the wide range of total point scores has been replace by a table. You look up the difference in the table and see how many imps you get. Part of the table is:

Points          IMPs  
20-40            1  
50-80            2  
90-120          3  
130-160         4  
170-210         5  
220-260         6  
270-310         7  
320-360         8  
370-420         9  
430-490        10  
500-590        11  
600-740        12  
750-890        13  
900-1090      14  
1100-1290    15  
1300-1490    16  
1500-1740    17  
1750-1990    18  
2000-2240    19  
2250-2490    20  
2500-2990    21  
3000-3490    22  
3500-3990    23  
4000+          24


The way the scoring works, if you are +10 imps on a hnad, your opponents are -10 imps. At the BBO, your imps are figured against each other table... and then averaged.

For instance if you are +100 and two other pairs with your hand are +!00 and -200. What is your imp score?

Your 100 versus the other 100 is a tie, for 0 imps
Your +100 verus -200 for the other guys with your cards is a +300 for you. Looking in the table, you see that +300 is +7 imps for you. So +0 +7 imps total 7imps/2 comparisons, your average score is +3.5 imps.

--------------------------------
And below is the body of the TEXT BY BBO GOLD STAR and favorite player, RADO... cut and pasted here for easy of acess

Will try to present the difference between Matchpoints scoring and IMP scoring and the different strategies on the table depending on scoring method

1. Matchpoints: your result is compared to the results of all the other pairs on the same board and is given some percent between 100 and 0
example: assume 5 tables (10 pairs) and N-S scored different +:
You've scored +140, 3 pairs +110 and only one +100
here you will get 100 % and your opps 0%, at the three tables where +110 scored all N-S and E-W pairs will get 50 %, the last N-Sfor their +100 only will get 0% while there opps E-W for -100 will score 100 %

at the above example there would be no difference whethere you have scored +170, or +420 or +1100

Simply the best result gets 100 % the next gets less.....

2. At IMP's your result is compared to the results of all the other pairs on the same board and is given IMP difference to all other pairs
for example: assume 5 tables (10 pairs) and N-S scored diff rent +:
You've scored +200, 3 pairs +100 and only one passed
here you will get (200-100) = +100 = +3 IMPS (see the table of Ben Inquiry) so you won by 3 imps to every of the three pairs and once (200-0)=+200=+5 imps another five imps from the fitht pair who passed out
IMP scoring is commonly used in teams matches:
for example you''ve made +620 for 4H just, at the other table your team mates sacrificed to 5C for -500 only the difference is +620-500=+120 = +3 IMPS for your team next board you go down one on 4H for -50, at the other tble opps bid only 3H and made 140, here you get -50 -140 = -190 = -5 IMpS for yoyr team.

the conclusion: when playing IMPS and you have safe way for 9 tricks at 3NT and risky way for +1 (but risking going down) you have to play safe and to get your +400 (or +600 vulnerable) because if trying for +1 and make you will win 1 IMP only but when failed you will loose 10 Imps ( 12 if vulnerable). Opposite when Matchpoints playing for +1 will give appropriate % result more (if all made +1 and only you just made then you will get 0%).

At Matchpoints reasonable risks for making overtricks can be accepted while in IMPS NOT.

[Chamaco]

Basically it boils down to this:
at any form of scoring, you want to take big risks only if you may have great rewards.

The amount of the reward is determined by the form of scoring.

At IMPS (or team match or rubber bridge), a big plus, such scoring as Game bonus, Slam Bonus or severely penalizing opps (+500-800-1100) will be cumulated over the boards, and will have a great impact on the final result.

At MP, it does not matter the amount of the bonus (or penalty) you score on the given board: what will be accounted will not be the magnitude of your score, but the number of opps who did better or worse than you.


So, at IMPS, you want to take risks to bid games:
you risk a safe partscore (say 140 points), in rder to go for the bonus (scoring 400/420 for NV game, 600/620 for a Vuln game).

So, at IMPS, you bid very tight games even if you think they are below 50% of success (usually around 40-45% NV, 35-40 Vuln) because the game bonus justifies the risk.

The converse applies to slam bidding:
it is true you have a slam bonus to shoot at, e.g. 6H = 980 NV (1430 vuln), but if you go down you will not score the points for 4H making ( 420 NV - 620 Vuln).

So the choice here should be more conervative, and you won't be bidding dubious slams unless you have well-founded chances to make.

Even more conservative should be the choice between small and grand slam: the extra bonus (500 NV, 750, Vuln) for the Grand is even higher but the potential loss of a cold small slam (980 NV, 1430 Vuln) is almost twice as big, so it is too dangerous to look for a grand slam which does not offer solid guarantees.

More points that should be dealt with talking of IMPS vs MP are:

1) preemptive tactics
2) penalties
3) defense issues (risking overtricks for declarer on opening lead or during play)

Certainly someone else can do it better than myself :-)

[Fluffy]

The main difference is these one:

At match point you focus on the maximum number of tricks avaible
At IMPs you focus on the exact number of tricks needed.

So at match points you just try for the maximum number of tricks, that means you have to try to make all the tricks avaible on each suit when it is divided as expected (3-2 normally), and safety plays hardly come into play.

At IMPs you focus on a goal of tricks, you shouldn´t care about overtrricks unless your contract is 100% assured (at part scores you can risk a bit more of course).

IMPs require that you count your winners before paying the first trick at NT contracts, there is no sense in trying a finese that will stablish a 7 card suit if you already have 9 without trying the finese and if it fails the contract can go down.

At suit contracts the proper way to count is not winnners, but losers, it means you stablish 1 hand (the one with longer trumps) as base hand and count wich tricks it is gonna lose, when there are more losers than you can afford you have to find a way to get rid of them (ruffing in dummy or discarding on dummies winners normally).

[Free]

The difference is not only in declarer play and bidding, but also in defending.

Sometimes you need a "magic card" in partner's hand to set a contract.
Even if chances are small, in imps you'll risk it despite the overtricks it might produce, to steal their game-bonus (or slam-bonus). The extra tricks you give them won't score huge, just a few imps (1 or 2 extra) perhaps. On the other hand, if you manage to set their contract, you'll steal their bonus, so you'll win 10-12 imps!
In MP's you won't risk it if the chance is too small, because if your partner doesn't have that special card, you'll get a bottom because of the overtricks.

Remark: this "magic card" is not an impossible card! If you know partner has maximum 1 HCP left, don't seek for the ♠A...

[mikestar]

Some critical bidding differences:

Game bidding: more aggressive at IMPs especially vulnerable.

Choice of game: always the safest game. For example if (hypothetically) you have a hand that is sure for 3NT on the nose but will make 4H if the trumps are 3-2, 4H is the correct MP contract but you belong in 3N in IMP.

Partscore battles: about the same, based on LOTT. You can even be a shade more agressive in IMPs vulnerable, as -200 vs -140 isn't as big a loss as in MPs and partscore doubles are more dangerous: If (both vul) they bid 3S over our 3H and every body else our way is in 3H +140, then the double is free at matchpoints.
-730 is the same bottom as -140 but +200 if they go down one is top vs. +100 for a bottom if you don't double. At IMPs the double isn't free at all you change a 7 IMP loss into a 13 IMP loss if you are wrong, and change a 1 IMP loss into a 2 IMP gain if you are right, so the odds against the double are 6 to 3.

Small slam bidding. Similar in both games. You won't win or lose big on 50% slams in either form of scoring.

Choice of slams: in matchpoints you try for 6N if feasible, especially if the alteranative is a minor suit. In IMPS always go for safety. For example, if 6D is laydown and 6N is 60%. bid 6N if you think most of the field will be in slam. (Safety is better in MP also if you think the field is missing this one.) At IMPS, 6N is idiotic.

Grand slams: In IMPs vulnerable it is only 17 to 13 against if the small slam is bid, but if they are stopping in game, you are risking turning a 13 IMP gain into a 13 IMP loss trying to get 17 IMP at odds of 26 to 4 against, it can't be right to bid. Conservatism is bidding grands pays well in MP too for the same reason. If the field are in 6S making bidding 7S is going top or bottom vs an average score is you stop in 6. All you need is a 50+% chance. But if the field is stopping in game, you are risking a bottom when you already have a top.

[inquiry]

I want to add this about part score battles..

Amazingly enough, it is "safer" to compete for part-scores in imps than matchpoints. At matchpoiints, you will be doubled a lot shooting for the magic 200 (or should be) in highly competitive auctions. At imps, doubling partscores (and thus mking it a potential game swing) is much more risky.

This doesn't mean you can be reckless in partscore battles. But if you are going down one at imps, odds are you will "never" be doubled, becasue the difference between +100 and +200 for the oppenets is not worth the risk of -670 or -770.

Another way to say this, is at matchpoints, most people need to be doubling more close part scores contracts, while at imps, most people are probably conservative enough

Ben

[jtfanclub]

Also, if I may interject (in spite of being an Intermediate)...

NT partials are very powerful in matchpoints. So powerful, in fact, that you often ignore 5-3 major suit fits to concentrate on no-trump (with otherwise balanced hands). The usual issue with no trump vs. a major is that they both make the same tricks most of the time, but when they don't the no trump gets set for a number while the major just goes down one (or has one fewer overtricks).

First hand: You 120, Field 110
Second hand: You 120, Field 110
Third Hand: You -300, Field 110

That's either -8 or -9 Imps, depending upon the scoring used. But it's 67% in match points.

Note that while 5-3 fits tend to play just as well in NT as the major (assuming the rest of the hands are balanced), but 4-4 fits play better in a suit,.

[Fluffy]

Competing in part scores because you aren´t gonna be doubled?

that doens´t look like a winning strategyat IMPS, compete too much and score -800 when cards are awfully and lose the match on a part score hand is not a good method. not to mention the usual fact of pushing them into makeable game (often helped by thinks and tempoes, but that is another problem).

Another thing...

I hardly believe BIL´s should care about the winning of bidding games vulnerable vs non vulnerable, that thing is often overvalued at high level. They have much more important things to care about than winning a 0.7 IMP or so in average on certain hands.