My choice of 1♦ in third seat is clear, a weak 2♦ opener would be a bit of an underbid with eleven points and a singleton. West's 2♠ call is a smart move which will cause N-S more trouble than 1♠ would, if the East hand is very weak and the North hand is a near opener. North makes a negative double promising a decent hand and four hearts, but I have no way to compete over 3♠, so it's defense on the first board of the day.

North makes the curious choice of the J♠ lead at my table, and I follow with the seven in the hope that partner will see this as a signal showing three trumps and a ruffing possibility. Declarer wins and takes the heart finesse, losing to my king. Before trying the 8♣, I cash the K♦. This allows partner to place the ace-king of diamonds in my hand, as well as the king of hearts, most of an opener already. With dummy so strong partner should be able to work out that I cannot hold many more high cards and still leave enough for declarer's bid. Next comes the 8♣, covered by West's king, and partner wins the ace, then cashes the queen and jack, allowing me to discard two more hearts. Partner leads a diamond to dummy's queen and my ace and I continue with the jack to kill dummy's ten. Poor West ruffs with the eight and partner overruffs! There is yet another trick available, a heart ruff, but partner tries a futile club and declarer takes the rest.

If something like this happened to you as West, my sympathies. I recall a hand from long ago where the opponents cross-ruffed the first eight tricks against me in 4♥, then led a card I could win. Somehow at this stage I wasn't able to claim yet and had to decide between two possible lines. The defender on my left said with a smile, "are there any matchpoints left?" Sometimes you just have to laugh and move on.

RESULT: 3♠w-3, 150 to N/S, 98% for N/S, 98% for McBruce after 1 board. (Shall I quit now while I'm ahead?)

Still nobody out there considering a weak two bid on an opening hand, I trust. It can happen to us all, especially playing with a good partner: we get excited and miss obvious bids or plays ... as I am about to. Partner's cuebid shows about 10 points in support of hearts, so 4♥ seems the obvious bid, and when dummy hits after the K♣ lead it seems a good decision. I win the A♣ and East plays the nine for some reason. This leaves me with one club loser and it looks like I can park the diamond loser on the K♠. I decide not to take the trump finesse and when East ruffs the K♠ I have two losers for a poor score, and a nagging feeling that there was no danger in the trump finesse. If it loses and West cashes a club I am no worse off than anyone else. My play would work only if West had a singleton or doubleton Q♥, probably not the most likely situation here. Oh well.

RESULT: 4♥s+1, 650 to N/S, 24% for N/S, 61% for McBruce after 2 boards.

I have switched to the West seat for this deal. After the 1♦ opener, the 1♠ overcall by partner, and South's negative double, I have enough to bid game: nine points and a key void, four-card support -- this is a game bid on power assuming partner's overcall isn't just fooling around. If I bid 4♠ right away and the opponents compete it will sound to partner like I have pre-empted. The SAYC solution: a cuebid. And not just any old cuebid: a jump to 3♦! Now when the opponents compete in diamonds I can pass and let partner make the decision. I don't know how good partner's overcall is, but partner knows I have a good hand in support of spades. When the J♣ lead is made (I have switched to the East chair; playing bridge against computers this is a tradition: human always declares so the computer dummy can go and drink liquid silicon or something...), I count five spades and five clubs and a diamond ruff for 11 tricks, two ruffs for 12. If the opening lead is from JT9x there is no easy way to make it, but clubs break and twelve tricks come rolling in.

RESULT: 5♠e+1, 680 to E/W, 52% for E/W, 58% for McBruce after 3 boards.

A most surprising result. 18 semibalanced points with a six-card club suit rates as a 19 count, and my jump rebid of 2NT shows 18-19. But the computer consensus seems to be a 3NT rebid, probably based on a simulation. Computers decide what to bid or play by rapidly dealing out the unknown cards and throwing out the distributions that don't match the clues given so far in the auction or play. Possibly the computer found almost no hands that responded 1♦ to 1♣ that would not give a decent play for 3NT.

With ten tricks in the minors (unless North has all four missing clubs), it is a contest for overtricks, and I am lucky enough to get a spade lead and a dubious play of the 9♠ by South, while others get a heart lead and a switch to spades for three defensive tricks.

Aside: Without looking at the other hands, how do you feel about the East-West heart stopper here? Is it solid against any lie of the opposing cards, or can the opponents cash the suit if they start from the right side? Clearly starting from the North hand gives West a trick with the K♥. If South begins the suit, West ducks and if North wins the Q♥, West's K♥ will force the ace on the next round and the J♥ in dummy is a stopper. So Jxx opposite Kx is safe. Clearly spades, JTx opposite Qx, has to be a stopper: but what about Jxx opposite Qx? Does this work or is there a possible weakness? Answer: As long as the opponents lead spades, you will have a stopper if you duck and wait for the other opponent's play. 3NT often depends upon the strength of declarer's stopper(s) in a weak suit. (That's why you should stay away from these recently-legalized notrump bids with singleton honours: just because it is now legal doesn't make it sound bridge.)

RESULT: 3NTw+2, 660 to E/W, 78% for E/W, 63% for McBruce after 4 boards.

Slaughter on Fifth Avenue! East has a very strong hand and 3♦ seems like a reasonable call. 3♦ becomes the contract at every table in the computer tournament, yet nobody can make it, and our result beats everyone! I choose the 6♠ for my opening lead, not wanting to imply an honour in spades by leading a smaller card. This goes to partner's jack and declarer's ace. A small diamond is led from declarer and I grab the T♦ to lead a second spade. (Declarer may have falsecarded with the AQ doubleton, but I'll take the risk.) Dummy's king wins and declarer pitches a club. Next is the 2♥, ducked by partner and declarer finesses the 8♥ into my king. I continue with a club and partner wins the king and ace, dropping declarer's queen. Declarer has an emerging problem, not wanting to have to lead trumps from hand, so on the next two clubs he pitches a small heart from hand and then the ace of hearts! Ruffing the heart continuation, declarer has no choice but to play trumps from hand, and partner gets two more diamond tricks for four down and a universal top!

RESULT: 3♦e-4, 200 to N/S, 100% for N/S, 70.4% for McBruce after 5 boards.

South doubles partner's response of 1♥ and I raise to 2♥, hoping to keep North out. North bids 3♣ and I have a decision to make. The Law of Total Tricks isn't much help, it says that assuming both sides have eight-card fits, that we can make 16 total tricks in our suits: if we make 3♥ (nine tricks) they should only have seven in 3♣, two down. If 3♥ is down one, so is 3♣. If 3♣ makes, 3♥ is two down. Perhaps I should double, since if 3♣ makes we are probably getting a poor score anyhoe. Trouble is, during the bidding it is hard to tell the length of the fits: either side may have a nine-card fit and the math changes. I decide to bid 3♥ and cross my fingers until the auction ends without a double.

Again the computer East goes off in search of computer nourishment while the human plays the hand. The A♣ is the opening lead, but then South switches to the K♦, which looks good until I realize the only justification for playing such a card: South must have a singleton K♦, and probably has three or four trumps. The ace and king of hearts reveal the bad break, and I decide on a small sacrifice as the best path ahead: I finesse the T♥, then play a small diamond to the jack, allowing South to ruff with the last outstanding trump. South cashes the K♣ and leads a spade to North's ace and I need the rest. I ruff the J♣ lead from North, cash the T♦, then lead the T♠ toward dummy's king-jack... North plays the queen and I am home! A look through the play reveals North's error: discarding a spade on the third heart. The difference is crucial, from an average board to a possible bottom for down one. Would I have finessed had North kept a low card to play? Probably: North has only shown the A♠ and the J♣ so far for his three-level call, and probably should have another high card.

RESULT: 3♥e=, 140 to E/W, 53% for E/W, 67.5% for McBruce after 6 boards.

OK, let's talk about psychic bids. Many new players feel that when a psychic bid works against them that they have been cheated. But psychic calls, as long as partner is as much in the dark as the opponents, are part of the game, just as bluffs are part of poker and razzle-dazzle is part of football. The reality is that most of the bad feelings that come from a working psyche stem from the victim's inability to adjust to clues about what's really happening. As you improve you begin to see when the cardplay seems off and if you follow the clues you won't be damaged as much by psychic calls. If the opponents have 16 points between them, RHO opened the bidding, and LHO turns up with the AKQ of a suit, you know that RHO's opener was not kosher!

On this hand, my 1♦ opener was made with the expectation that I would jump rebid 2NT and play 3NT, and possibly avert a diamond lead. Things went off the tracks when LHO overcalled 1♠ and RHO bid 1NT! I doubled, which the computer read as takeout and a fairly good hand of 16+, and the hands that my computer partner came up with that fit the auction so far told an interesting story: that passing the double was likely to be best!

In fact, it wasn't; with 3NT on, our side needed to get this down three to beat the pairs in 3NT. We might have if I had switched to the Q♦ at some point, so that partner could start clubs. Instead, after establishing two spade tricks I played ace and another club, and we got two spades, four clubs, and two hearts, declarer winning the K♣, the A♦, A♠, and two hearts for down two.

My diamond opener kept the opponents from finding their diamond fit, and few N/S pairs got to 3NT. So my psychic call worked. Next week it will be partner with the five small diamonds and I'll be raised to five for a bad score...

As a Director, I can tell you that what we look for after a psychic call is bids and occasioanlly plays by the partner of the psycher that indicate some suspicion of what is going on. Failing to raise or lead the psychic bidder's suit, deciding not to double when partner is ultralight for his actions, that sort of thing may lead to a score adjustment or a stern warning. But there is no infraction in deliberately misleading the opponents, as long as partner is equally misled. That's what psyches are meant to do: their success rate is random, and probably worse than not psyching in most cases, but nobody tends to notice when they don't work. What you should do when a good player's psyche victimizes you is smile: if the good player feels he needs a psyche to beat you, you must be improving!

RESULT: 1NT*e-2, 500 to N/S, 88% for E/W, 70.43% for McBruce after 7 boards.

The computers think that the West hand is an 1♥ opener, and have the good sense to rebid 4♥ (promising a minimum opener without a singleton or void) after partner makes a 2NT Jacoby raise. I pass and allow partner to open 1♣, and when my 1♥ response is raised, I re-evaluate my hand. The QJ doubleton of clubs is pulling a lot more weight after partner's opener, and I bid 4♥. So far so good. But in the play I lose my mind. I win the Q♦ lead in hand and play a low heart to the king and South's ace, noting North's play of the nine, but not noting it as well as I should. The play of the nine makes all of West's trumps good after the ace is played, but more importantly, it means that if North's diamond lead was from a five card suit, giving South a doubleton, I can ruff two diamonds in dummy without fear of being overruffed, since the lowly five-spot is as good as the queen!! My play at trick three should be a low club. If South wins and plays a second trump, I can still ruff two diamonds in dummy and pitch a spade on the third club. Instead, I pull a second round of trumps; fatal to my chances of making the crucial overtrick. I cash the ace and ruff a diamond, then lead a club to the jack, South ducking. Another diamond ruff follows, but when I play another club, South wins and puts me in dummy with a spade. My third club, on which I should have pitched a spade loser, is ruffed by South, who has tossed away two clubs as I ruffed out the diamonds. This forces me to overruff and the spade loser is still there. Ten tricks when the rest of the computers are making an easy eleven is not a recipe for success. Even worse is that the 200 computer players at the other 50 tables finished last night as I was sleeping and are probably watching, shaking their connectors in astonishment at the gaffe of that silly human player...

RESULT: 4♥w=, 420 to E/W, 0% for E/W, 61.63% for McBruce after 8 boards.

Not much here; every computer table had the same auction I did, and all led a small club; I chose the 7 instead of the 5, because I like MUD (middle up down, leading the middle card from three small) against suits, as long as the middle card is not too low as to suggest that you are leading from an honour. Let's be clear here, leading from three small is not usually your ideal lead, but it beats leading away from the queen of trumps or a side suit ace. We get our two clubs, a diamond, a heart and a trump...and this is somehow a fairly good score, with nine declarers finding a way to make nine tricks. East deciding to lead a trump at some point? I can't see another way, even cashing the ace-king of diamonds doesn't give declarer enough pitches on the established diamonds to avoid the normal losers. Mystery.

RESULT: 2♠s=, 110 to N/S, 59% for E/W, 61.33% for McBruce after 9 boards.

A strange quirk of standard bidding here leads to me and all 50 computer Souths opening this hand 1NT, not 1♥. The reason: no good rebid should partner respond 1♠. After 1♥ (pass) 1♠ (pass), what next? 1NT shows 11-14; we have 16. 2NT shows 18-19; we have 16. 2♥ promises no extras; we would open this hand if the ace of hearts was instead the two of hearts: extras. 3♥ shows a good six card suit; we have five only. Raising to 2♠ or 3♠ is dicey since partner may have four small spades. So 1NT is the correct opening call, and when it goes down while hearts makes nine tricks, we shrug and move on: what else can you do? By the way, when you see such a dummy and know you are in the wrong contract, don't give it away. Say "thanks partner" and call for the diamond from dummy. Exude confidence. Perhaps one of the defenders will convince themselves to switch to a different suit. Perhaps they will block the diamond suit. It wouldn't be the first time.

RESULT: 1NTs-1, 100 to E/W, 56% for N/S, 60.80% for McBruce after 10 boards.

Did you know that the takeout double actually predates contract bridge? Our game in its current form started with a cruise through the Panama Canal in 1925 with Harold Vanderbilt, a dozen or so copies of his new scoring table that is largely unchanged today, and some friends (one female kibitzer whose name was not recorded for posterity is said to be the originator of the term 'vulnerable'). The takeout double is an invention of the pre-World War I bridge-whist era. Over 100 years later we're still working on the details.

This North hand is not exactly textbook for the red card, until West opens 1♣. Now North has no convenient call: no spade or club overcall describes the strength of the hand, and a cuebid shows both majors. The takeout double is the least worst call. East bids 1♦ and West bids 1NT, which North can almost certainly beat, but settling for +100 in 1NT doubled doesn't seem sufficient here. North's 2♠ call after making a takeout double shows a very strong hand, and South is taking the partnership into considerable danger by passing 2♠ with three card support and two doubletons.

And yet, the defense holds this to nine tricks! East leads the six of clubs, and West puts in the nine, which wins! The queen of clubs from West is also ducked, for reasons I cannot fathom. The computer also ducked the king of hearts at trick three and the king of diamonds at trick four, winning the rest for a complete zero. If you can justify this play by the computer, let me know. I think the machine was throwing a tantrum after South parked the car in 2♠...

RESULT: 2♠n+1, 140 to N/S, 100% for E/W, 64.36% for McBruce after 11 boards.

After what seems like the simplest of auctions, I take over the North seat to declare as the computer player goes off to do whatever computers do when they take a break. The 6♦ is the opening lead and prospects look alarmingly good. Have we missed a game here? It looks like two hearts and a club may be our only losers in 4♠. The North hand has fifteen points and two doubletons, and at many tables, one would expect North to make another call over the raise. 4♠ may fail if trumps break badly or if there is an unexpected ruff. But it looks like I am set for a poor result here. I win the diamond in the North hand and play the A♠, noting the ten from West. This could be a singleton, or a doubleton JT, or possibly even a falsecard from JTx. If it is not a singleton, the normal play of trumps from the top will work and those in 4♠ will probably succeed, so I take a chance and play a spade to dummy's 8 next, and West discards! I next cash the Q♠ and play the K♣. East wins the A♣, cashes the A♥ and continues with another heart to West's king. West leads the Q♦, which should be the fourth trick for the defense when East ruffs, but East seems to think there is a chance to hold me to eight tricks by not ruffing yet, and instead pitches away a club. This is fatal: I win in dummy, ruff a heart to my hand, pull the last trump, and play the rest of my clubs. East's discard on the Q♦ lead has made the 4♣ good for a tenth trick and a huge score. This is surprisingly bad play from the computer, and I wonder if I haven't weakened the computer skill level a bit too much...

RESULT: 2♠n+2, 170 to N/S, 98% for E/W, 67.17% for McBruce after 12 boards.

In around 1922, the manager of baseball's Detroit Tigers received a letter from a fan in Iowa, who claimed that if he was sent the $4.75 train fare he would come to Detroit and strike out Ty Cobb on three pitches. The manager figured there was nothing to lose, sent the ticket and arranged for the young prospect to come into the big city on an off day. After some warmups, the superstar Cobb grabbed a bat and assumed the position in the batter's box. Cobb hit the first pitch over the left field wall, the second pitch over the right field wall into the street, and the third pitch to straightaway center field, nearly reaching the upper deck. The manager walked to the pitcher's mound slowly and said "well, kid, what do you have to say for yourself?"

The kid replied, "I don't believe that's Ty Cobb there."

Well, after watching the computer happily hit three diamond tricks out of the park against 6♠, what I have to say for myself is, "I don't believe that is a 3♠ jump rebid there." If the jump to 3♠ asks me to bid game unless I have a minimum 1♠ response, and I have a full 13 plus shape, we should be in the slam zone. We're not because North doesn't have enough for the 3♠ call. That said, I might have done better by bidding 4♣ over 3♠, which would certainly get a 4♦ call from North with a diamond control. Even the king of diamonds at least gives the slam a fighting chance.

RESULT: 6♠s-2, 200 to E/W, 0% for N/S, 62.00% for McBruce after 13 boards.

It's well-understood by most newer players that the way to show a limit raise or better in partner's suit after an opponent overcalls is to cuebid the overcalled suit, as in auctions like 1♥ (2♣) 3♣ or 1♠ (2♥) 3♥. In this one case, it is okay to have three-card support: limit raises in uncontested auctions should always be four-card raises. But when the intervening bid is a takeout double, things are different. Redouble, you might think, would be the logical way to show a raise, but in fact redouble has evolved to the message "no fit, but cards enough to ensure that it's our hand, don't let them steal it away." To show a fit with some values after an opponent's takeout double, the standard call is 2NT. This is different from 2NT in an uncontested auction, which is forcing to game and promises four-card support. After a takeout double, three-card support is fine, and ten or more support points (counting shape if needed) are sufficient.

Baffling to see most of the field finding a third trick on defense. Who switches to a trump at trick two from Kx? It works, but it's not a play many are likely to find.

RESULT: 4♥e+1, 450 to E/W, 20% for N/S, 59.00% for McBruce after 14 boards.

Easy to see after the play is over that the winning action here, after 1♣ (3♠) Pass (Pass) is a simple takeout double, which finds the magic heart fit and scores all the tricks and all the matchpoints, as long as they don't find the diamond lead. Problem is, on a bad day partner will bid diamonds, not hearts, and will insist on them several more times, eventually giving up a "telephone number" (bridge jargon for going for 800 or more) when you have only three-card support. On a really bad day partner will decide that the double is for penalty and the opponents will make nine tricks for +530. Rebidding 4♣ is probably best and certainly safest, and from that point on it is a Clint Eastwood "do you feel lucky?" auction. A lot of cohones in the computer tournament; might be interesting to see how the actual humans handle this deal...

RESULT: 6♠*w-3, 500 to N/S, 88% for N/S, 60.93% for McBruce after 15 boards.

I see no reason not to open 1♥ with the West hand as dealer. I should be able to rebid 2♥ over a 2♦ response and 2♣ over almost any other forcing call. If you choose not to open 1♥ it seems silly to pass as the computers all did, when 2♥ is a reasonable second choice if you feel 1♥ is too much. Players seem to have this idea that between a one-major and a weak two there is this small chasm of hands that are unsuitable for either. I have found that passing these hands usually gives you an even harder decision on the second round. A look at the six-packs shows what passing gets you. Partner opens 1♠ and you haven't enough points for a 2♥ response so you have to bid 1NT with a 6-5 hand. Sure enough, partner bids 2♦ and who knows what is best now? Grab the bull by the horns: bid 1♥ to begin with! Even when partner insists on notrump, you can complete your picture with a 4♣ call, and let partner make an informed decision.

Another strange (dare I say human?) defense by the computers, beginning with the J♦ opening lead, won in dummy with the ace. South grabs the ace of clubs at trick two and tries to cash the K♦, which is a disaster: declarer ruffs and a club ruff gets declarer to dummy where the Q♦ allows the T♠ to be discarded. When the K♣ falls in three rounds, West has eleven tricks on a crossruff, losing only to the A♣ and the ruffed Q♦, a result which is unmatched at any other computer table.

RESULT: 4♥w+1, 650 to E/W, 100% for E/W, 63.38% for McBruce after 16 boards.

Did you respond 2♣ over partner's 1♠ opener? It's an easy trap to fall into. On this hand it won't even hurt much. But on another day, this will lose a heart fit. 2♥ over 1♠ promises, 100% of the time, a five-card heart suit (or longer). You can't do it with four hearts, no matter how strong you are. You'll always be able to bid a four-card minor or support spades with three, and if you bid a four-card minor, partner will bid 2♥ with four of them, so you'll never lose a heart fit by that route. Now, if 2♥ promises five, it logically follows that if you don't use that rule and bid 2♥ over 1♠, partner will have a hard time believing you have five hearts later in the auction. By bidding 2♥ you may have trouble convincing partner your clubs are as strong as they are, but nobody loses sleep over missing a fit in clubs. Anyhow, on this hand, most people will end up in 3NT unless you decide that the misfit is worth a significant point deduction. The devlish East at my table found the inspired lead of the 4♦ and West played along and ducked the king, allowing my nine to win the first trick but convincing me that East had the K♦ and preventing me from even hoping for an overtrick. But with clubs destined to produce five tricks however they were played, the contract was safe enough for a decent score.

RESULT: 3NTn=, 400 to N/S, 56% for E/W, 62.94% for McBruce after 17 boards.

It seems like an auction from deep space, but is duplicated at all 50 computer tables. West has that type of hand and the vulnerability demands a sacrifice. Our poor result was due to the strange choice of opening lead, partner leading the 3♠. Normally we would get three trumps, two hearts and the K♣, but this uninspired lead allowed declarer to win the A♠ and pitch a heart loser from hand. The lesson for this hand is not "don't lead a spade." The real lesson is "when partner makes an error that is clear to everyone by the end of the play, it cannot possibly help to point it out.". Make a joke of it and move on, rather than restating the obvious. Partner has known for thirteen tricks that the opening lead sucked, why amplify that? Part of the game is getting partner to play good bridge. Few people can do that when they are constantly criticized.

RESULT: 5♦*w-3, 500 to N/S, 9% for N/S, 59.94% for McBruce after 18 boards.

Not sure why the program placed me in the West chair for this one, no real decisions to make. Nothing to suggest anything other than pass cards in the auction, and East ran the defense, leaving me only to follow suit and cash tricks when in. We got an extra trick worth about half a board when North inexplicably pitched a club from hand on the second round of spades, apparently fearing losing trump control. Other than that, not much here. I note that hands where nobody has an eight-card fit in any suit seldom seem to be a good result for the declaring side, no matter how quickly they give up and pass. Not sure how to use that information in the bidding though. Almost all systems are, for good reasons, consumed with finding a fit early, not with finding that there isn't one.

RESULT: 2♦n-3, 150 to E/W, 85% for E/W, 61.26% for McBruce after 19 boards.

Twelve tricks are there if the spade finesse works, but two down when it doesn't. East made a good switch to hearts after winning the ace of diamonds, ensuring a third defensive trick. If I pass 4♠ and make four I get 27%, 75% if I make five (maybe in 4♠ they don't switch to a heart at trick two). Let's be generous and say I'm getting 40% in 4♠ on average. So bidding slam improves my score on this board by 60% half the time and reduces my score by 40% half the time. Those are good odds! If we get only 30% on average in 4♠, the gain is even better, +70% vs -30% with equal chances of each. It seems right to bid this based on the potential gain, even though the operation was not a success in the end. Matchpoints is a game where steady play gets you only so far, and sometimes you have to take a few risks to move from the 53-55% range to the winner's circle.

RESULT: 6♠s-2, 200 to E/W, 0% for N/S, 58.20% for McBruce after 20 boards.

I fumble this one. East hands me a free trick by leading a small spade against 3NT. The computer would have made eleven tricks yawning but I manage only ten after switching to the North chair to declare. I lead a diamond at trick two, intending to cover whatever West plays, finessing the nine if West plays low; West in fact plays the king, so I win, lead a club back to dummy, and shoot a second diamond through. West takes the queen and I somehow fail to notice East's discard. I cannot even praise East for playing a card of the same colour, for East's discard is the 9♥, after I played the 9♦, so I have no excuse here. Under the mistaken belief that my diamonds are ready to run, I win the spade return and play the Q♥ to knock out the ace, hoping for eleven tricks. East ducks and I count eleven tricks, so I switch back to the J♦, then a club to the king, then the 8♦. When this trick ends I click on the 7♦ only to discover that West has already led to this trick, having won the previous trick with the ten! The ace of hearts is the crucial third trick for the defense, and I have earned a bad score. Three spades (after the spade lead), three hearts (after the ace is knocked out), the ace of diamonds and four clubs (when clubs break 3-3) is the normal route to eleven tricks. Even without the spade lead, you can force a second diamond trick once the ace of hearts is dislodged. My bad.

RESULT: 3NTn+1, 630 to N/S, 9% for N/S, 55.86% for McBruce after 21 boards.

Several teachable moments here, possible mistakes for new players. Did you as South consider opening 1♣ and reversing to 2♦? A reverse (an opener's rebid that forces partner to the three-level to prefer opener's first suit) does show extra values, about an ace more than a minimum opening bid. You have that, change either ace in the South hand to the three of the same suit and you still have an opener. But the other, equally important rule about reverses is that they always, always, show more (never equal) cards in the first-bid suit. The correct opening bid here is 1♦ and your rebid will probably be a jump to 3♣. Except that partner bids 2♣! There is a lot to be said in matchpoints for simply ending the auction in 3NT, which should make most of the time: partner must have something in the majors to be able to make a two over one response. But 6♣ looks like the place to be, and we can get out in 5♣ if partner is aceless, so Blackwood is the next step if you're looking for excitement. Partner shows one ace and my ticket to excitement is booked. I even get to be declarer, switching seats with the computer partner to attempt this slam finale.

There are two leads that kill off 6♣ before declarer plays a card: the A♠ and the T♦. But who can blame East for not wanting to lead an ace on this auction, nor wishing to lead a singleton when ♣QT doubleton may be a natural trump trick, or even a sneaky one. (Try dropping the queen the next time you have this holding and your RHO has the ace and king. You'd be surprised how often they finesse the next round into your ten!)

East's passive heart lead leaves declarer only one chance: win the Q♥ in dummy, drop the Q♣ in two rounds, cross to the J♣, pitch two spades on the ace-king of hearts, then take the diamond finesse. It fails, and diamonds break 4-1, but there is a trump left in declarer's hand to ruff the fourth diamond, and trumps left in dummy to ruff spades. Twelve tricks are there for a tied top!

RESULT: 6♣n=, 920 to N/S, 99% for N/S, 57.82% for McBruce after 22 boards.