[tgphelps]

My wife/partner and I are playing in a Swiss event on Sunday, and we've never played IMPs before. (We have a total of about 40 masterpoints between us, if you're scoring at home.) I've been googling for IMPS strategy, and haven't found much more than this:

https://kwbridge.com/imps.htm

If you have pointers to more or better information than is on that page, I'd love to read them.

One specific question I have is: Is the Law of Total Tricks just as applicable at IMPS as at matchpoints?

[P_Marlowe]

Hi,

The LoTT is relevant for IMPs, in the same sense, it is relevant for MP.


The simple answer is, bid games, ..., if you think of accepting an invite,
just bid it.

BOLS tip, especially relevant for IMPS: The 5 level belongs to the opponents.
Or in other words, it is usually best, just to take the offered money.
A sac. bid 4S over 4H is usually ok, the famous transfer, and a followup of the
LoTT, but bidding 5 over 5, is usually not.
At MP you have to get it right.

With kind regards
Marlowe

[mycroft]

The key - mentioned in that article, I'm sure, because it is in all, but maybe not as clearly as it could be:

• At matchpoints, it doesn't matter how much you win or lose by, just how many you beat.
  
• At IMPs, small wins aren't worth much. Big losses can kill the match. (of course, big wins can save the match).

I have definitely lost a match where we scored IMPs on 5 of the 7 boards, but the one loss was 13.

So (as all the articles say):

• Play safe. Don't push for overtricks at the risk of the contract, don't overcompete, don't double for a 1-trick set.
  
• Go for the game bonuses. Don't worry too much about conceding 4♠+1 if there's a line that might set them.
  
• But because they won't double for a 1-trick set, and because a double partscore swing is the same as a NV missed game, as long as you "are reasonably certain" you won't go down more than one, and it might push the opponents to a contract you might set (not game), go ahead and compete.
  
• ...

Another thing that is true and obvious from all the points the articles make, but may not actually be stated: Each board is worth the same at Matchpoints. The worst you can lose by a bad mistake is 4% of your score (2% if it's going from "average" to "zero"), and you can't win an event on one board, either. But you have to worry about every hand, because the overtrick you lazily concede on board 4 is the same 2% as your wrong-time interference over 1NT going for 1100 is. But at IMPs, some boards are worth (much!) more than others, and it's important to recognize that.

Having said that, to an extent, IMPs is easier than MPs. You just play the contract bid - try to make it, try to set it. The first thing you have to do at MPs when dummy comes down is work out what the contract *actually is* - "Oh, everybody's in 3NT, I have to try to make 4 (or 5) to get a good score." "Oh, lots of people will be in 2♠ instead of 1NT, we need to make 120" (but of course, the opponents might compete over a major fit, and the "lots of people" will be in 3♠ which probably goes down, so maybe 90 is lots better than -50?) This is even harder on defence. At IMPs, that all goes away: you know your goal; defeat the contract.

The other thing you have to handle (most of the time, anyway) is the fact you have teammates. At pairs, all you have to manage is your partner's wishes. At teams, you will get the feeling that you're letting them down (or you might get the feeling that they are dragging you down). Start saying to yourself now: "We win as a team, we lose as a team, there is no us and them". Try to believe it. Definitely act like it at the table, and at the score-up. Your teammates (and you) will play better for it.

[jillybean]

Good luck, have fun.

You may find this useful, written for our local unit.

If you haven't tried teams before, here's a guide on some key
differences between a teams event and a matchpoint pairs event.

The first is you have a team of 4 consisting of a N/S and E/W pair. If you need a pair for your team,
sign up at the partnership desk at a tournament. At the start of the game, you will be assigned a team #
and a "home" table, say B2. Your entire team sits at table B2, until the director says to start the match
and move. The N/S players remain at table B2. The E/W players sit at the "corresponding" table.
Usually, the "corresponding" table is the same table number with an adjacent letter. So the E/W
players go to table A2. The director will advise if there are any exceptions to this.

The second is that each board you play is not compared across the "field". Rather, it is only compared
with one other team. So, if you and your partner are sitting N/S against team #3, say, then your
teammates are sitting E/W at another table against the team #3 N/S players. After you play some
number of boards against the other team (typically 5-7 boards), the E/W players return to sit at
their home table with their N/S teammates. You then compare the results of each board played at both
tables.

The third difference is the score you win or lose on a board depends on the magnitude of the difference
in results. For example, if at your table you bid 3S and make 4, then your result is 170 on the board. If
the opponents at the other table bid 4S and made 4, then your partners have a result of -420 or -620,
depending on the vulnerability. So the net difference on the board is -250 or -450. This highlights an
important feature of team scoring: if you miss bidding a game that makes, and the opponents bid it,
then you lose a lot, especially if you are vulnerable. It is important to avoid large negative scores in
team games (or "telephone numbers", like -800 or -1100), unless you are very confident your partners
will be getting even larger positive scores!

The net difference on a board is translated into an "IMP" score (short for International Match Points),
using the IMP table you see at the side or bottom of most scorecards. So the difference of -250
translates to -6 IMP's, and -450 is worth -10 IMP's. For each board played against the other team, you
calculate the net score difference and convert that to IMP's. Note that a difference of 0 or 10 is
considered a "tie", or 0 IMP's. So another feature of IMP scoring is that there is no IMP advantage to
scoring 430 over 420, which is significant in matchpoint scoring.

You total the IMP's for the boards, both positive and negative, and see the net IMP result. Typically,
the team "captain" that won the match (with a positive IMP result) goes to the other table to confirm
the IMP result. If there is a discrepancy, the captain should check each board with the other team to
find the discrepancy. Once the result is agreed upon, the winning team captain completes a little card
that shows the winning team's number, the losing team's number, and the net IMP result. Both team
captains then initial the card, which the winning team captain delivers to the director. Note that there
may be a tie, where the net IMP's are 0, in which case either team should complete the card and report
the result.

The director compiles the results of the matches, and then posts the table assignments for the next
round of play. The team captain looks for the table assigned to their team number, and directs their
N/S pair to that table. Let the next round begin, and have fun!

[mikeh]

I don’t disagree with anything I’ve seen in this thread, so my post may be largely redundant but:

1. No need to stretch for a thin non vulnerable game, but definitely stretch for thin vulnerable games (don’t go nuts, though….the imo odds that merit stretching for vul games are based on the assumption that you will rarely be doubled and rarely be set more than one trick. If the auction suggests bad breaks, be a little more cautious unless you are sure of a good fit in trump (either because you have great, long trump or because partner has shown a good fit).

2. When declaring any contract, forget overtricks unless you are 100% certain that playing for an overtrick or two CANNOT jeopardize making your contract.

3. If your contract looks very bad, try to imagine a lie of the cards that might let you make it. That’s easier said than done but it’s worth taking time at trick one to formulate a plan. But…if you’re doubled and playing to make risks multiple undertricks, consider minimizing your losses rather than playing to make IF the auction and the play so far indicates that you are VERY unlikely to make…going 800 on a hand where you could have got out for 200 or even 500 is expensive. Similarly, if you are defending and it looks as if they are going to make….consider if there is a plausible lie of the missing cards that could result in a set, even if playing for it probably costs an overtrick or two. Giving up 650 when you could have held it to 620 is one imp. Beating it is 12! Similar calculations, with differing odds, apply to all contracts.

4. Don’t try to win the match. Play each hand without worrying about other hands, including a disaster. Many years ago I used to play on teams that weren’t as good as the ones I play on these days. I can’t count the number of times teammates would have a disaster early in a match and felt that they had to ‘get it back’. The usual result was two more disasters….and losing a match we had won because at our table our opps also had a disaster or two. Lose a match on your disaster? It happens. No pair…not even the best pair in the world..escapes disasters all the time. If you have one or two…hope your opps in your seats at the other table duplicate your result. This was my approach back when Swiss was scored win-loss. These days it’s usually victory points and that makes this advice even more relevant…lose matches by 5 imps or less and you’re still going to do ok if you win half your matches. Lose a few by 30+ imps, kiss the event goodbye regardless of which bracket or strata you are in. So don’t try to get bad boards back. As I say to less experienced players….’so you had a disaster….that’s what teammates are for’.

5. Don’t double partscores (for penalty) if making gives them a game bonus. The only exception is when your hand tells you they are way overboard….never….never….double for a one trick set. Never double only because ‘we have most of the hcp’. Any xperienced player can tell you of hands on which they or their opps ‘bid and made’ game or even slam with a combined 15 hcp! (The first time we did it, we bid 6S as a sacrifice and nobody was more surprised than us when it was cold)

6. Focus is critical at all forms of bridge but at imps you actually have it relatively easy. As dummy, just relax. I used to follow the play closely, trying to reconstruct the three hidden hands, thinking that it would improve my defence and my declarer play, where one is constantly trying to picture the two hidden hands. Nowadays, I think that mental energy is too good a resource to squander it when dummy….relax. Similarly, say you are declaring a cold 3N….you have an easy 9 or 10 tricks….ok, if you’re feeling really focused, by all means consider a (100% risk-free) play for an extra trick but don’t sweat it. Yes, you can lose matches by an imp, but (at VPs) it’s rare that a 1 imp loss here or there will affect your overall results so save your energy for difficult contracts

[P_Marlowe]

 mikeh, on 2024-August-09, 12:00, said:

<snip>
6. Focus is critical at all forms of bridge but at imps you actually have it relatively easy. As dummy, just relax.
<snip>

I think this is one of the most important hints.
You can also relax, if you see they are in game with 30+, and for whatever reason declarer tries to enginieer an add. overtrick
trick.

[helene_t]

Above all, don't overdo it. You will do ok if you use the same strategy as you are used to:
- the differences between IMP and MP strategies are small. It's mostly about borderline decisions which don't appear often, and since it's borderline decisions it will rarely be terrible to apply MP strategies at IMPs
- Especially if you are used to play in a weak or heterogenous MP field, you probably use IMP strategy already as you will be used to play to make your contracts (if a substantial part of the field is in a different contract, overtricks probably won't matter much)
- You will do worse if you let the scoring issue distract you from your play and bidding
- You will confuse partner if you suddenly try to adopt a new style

[tgphelps]

You all have been a big help to me. Thank you. I will read your replies again, and show them to my partner.