take a trick with a ‘next’ ace when partner ordered up
from third chair, and then lead trump to his partner.”
Seemingly always, but not always, I imagined.
The fact that Champ’s ace is a singleton enhances not only its likelihood
of flying, but also the likelihood the partner will have a card of that suit
to lead back at the end of the hand.
That’s an argument for “Lead trump, damn it!” as Ryan Romanik likes
to say.
But Champ’s ace lead was a concept worth testing; so Todd ran off to test it
with Fred Benjamin’s simulator
while I ran “hands on” tests in Gerry Blue’s Euchre Laboratory.
Todd ran thousands of hands of various configurations with the simulator, and I ran
hundreds of the same configurations in the Lab to look for patterns and reasons
for the simulation results.
First we gave the player in third chair three tall trump, since he ought to have
that much to order from third chair.
My suspicion after viewing just a few hands in the Euchre Lab was that Champ
was getting away with his ace lead, but not optimally profiting by it.
The early simulations tended to confirm my suspicions.
We chose hearts for trump (ordering a queen):
Right-king-ten, then left-ace-ten, in third hand.
We gave a small heart and a “green” singleton ace to first chair (we
used a club for consistency), then a “next” singleton ace (diamond), then
two aces (one of them – the club – a singleton); then we played with
a three-suited first hand, a four-suited first hand; etc., etc.
The simulator consistently produced both more marches and more euchres than I
found in the Lab, and the reasons appeared to be that the simulator would not
screen out (1) hands on which the dealer’s partner would “assist,”
pre-empting an order from third chair, (2) hands in third chair good enough to
go alone on, and (3) hands in which two diamonds wound up in third chair, giving
the player there an incentive to wait for “next” rather than order.
But we adjusted for all that.
The test was bifold, in both the Lab and the simulator:
We compared an opening trump lead to an opening ace of clubs lead (followed by
a trump if the ace flew). Early
tests, with two aces in first chair (one of them a singleton), favored leading
trump over leading the singleton ace – in net points, in most marches, in fewest
euchres. Then we tested hands with
only one ace (still a singleton).
When the ace was “next” (the diamond), the results were too close to call.
Then we went back to “green” (ace of clubs), threw in at least one low trump
(always) and sometimes a “next” card, and filled the rest of the hand with
the other “green” suit – two or three spades.
Net points then consistently favored the singleton ace lead, and we began seeing
a lot of euchres on trump leads.
We changed the third chair holding to both bowers, the ace and queen of spades
and the nine of clubs. The euchres
continued.
After a little study we figured it out.
In the first place, you have to expect the dealer to discard a spade or a club
when a queen of hearts is ordered to him (that gives him a better chance to trump
the opening lead, which is more likely to be “green” than “next”
if it’s not trump). And then it
became clear that, with no more than two black cards in the dealer’s hand
(and not likely spades, considering what the maker and his partner had), the
only way to get euchred, with this configuration, was for one of the opponents
– the dealer or his partner – to hold three (or four) trump with
diamonds to run once he caught the lead on the third trick.
And he did not need two diamonds to run if his partner had two and could overtake
the first diamond. The key for
the defense, once the lead was gained, was to avoid leading a club or a spade,
as both of those suits were controlled by the makers.
Another key – and this may be something new – is that the partner of the
opponent with three trump must keep one suit, to the derogation of others,
ready for the run. For example,
let’s say the dealer had three trump (enough to gain the lead on the third
trick) and a diamond to lead to the fourth trick.
His partner might hold ace and nine of diamonds and queen of clubs down to the
third-trick squeeze. Partner must
throw the queen of clubs in favor of the nine of diamonds in this situation,
because it is the run that produces the euchre.
You would save the queen of clubs with the idea of stopping a march, but it
won’t work for a euchre. (Todd
and I are considering patenting this ploy.)
But another pattern developed in the first chair holding of ace of clubs, ten of
hearts (trump), and three spades (king, ten and nine) – whether the partner
in third chair ordered on two bowers and ace of spades, or on three tall trump.
Although leading the ace of clubs would result in fewer instances of being euchred,
leading trump would produce significantly more marches.
Just like drinking Postum, there was a reason the ace of clubs lead prevented
euchres. As we said, the opponents,
to euchre, must hold at least three trump in one hand, and diamonds to run at
the end. If the ace of clubs is
led, it forces that three-trump opponent’s hand to ruff, and then he
no longer has enough trump to regain the lead to cash the diamonds.
But it’s that ruff also that would stop a march that was in the cards.
The differences were striking. In
a number of 1,000-hand samples, the euchres on the trump lead exceeded those on
the singleton ace lead by as many as 100 to 140, or 10 to 14 per cent (singleton
ace euchres ranged from only 10 to 47 per thousand, compared to trump lead euchres
ranging from 100 to 190 per thousand).
And marches on the trump lead exceeded those on the singleton ace lead by as many
as 240, or 24 per cent. But the
extra point gained on a march is only one-third as valuable as not getting euchred
(a euchre is a 3-point turnaround – the point you don't score plus the 2
points the opponents do score).
Thus 100 euchres, on the low end, constitute 300 net points lost compared to
only 240 net points gained by marching on the high end.
And thus net points favored the ace lead.
(Mind you, the results stated in the previous paragraphs are not general averages.
We were not seeking averages of marches or euchres in this experiment on a trump
lead vs. an outside ace lead – and you can't, given the idiosyncrasies
of different players you will find in third chair.
The figures in the previous paragraphs are merely examples from batches of runs
on a few finitely defined hands in which the outside ace lead works better, for
limited purposes, than a trump lead.)
So, students, here are the conclusions (take notes now; this will be on the final
exam): “Lead trump, damn
it!” still is the best policy in general.
But not if you have only one ace, and it’s a singleton, and it’s
“green.” In that case
you’ll make more net points leading the singleton ace.
You’ll gain fewer marches that way, but you’ll get euchred less often.
So, one more refinement: If you
have the lead and 8 points, go for the march.
Lead trump.
But if your opponents have 8 points, avoid the euchre:
Lead your singleton ace.
Got it?
If not, just lead trump, damn it.
And there may be another lesson here:
Keep your mouth shut in third chair unless you have a sodbuster.
Champ is on to something – but not everything.
After we ran all these tests, I got another message from Todd:
“Monday, Champ was sitting in first holding jack of diamonds, ace and queen
of hearts, ace of spades and nine of clubs; and he ordered up the ten of hearts.
And led his ace of spades! Which flew!
Then he led the left bower, which took the trick, everyone following suit.
Then he cashed his ace and queen of hearts and finally led his nine of clubs, for
which I had saved the ace for a stop.
I ran that hand through 500 hands in the simulator.
The ace of spades opening lead resulted in a net of 339 points, and a left bower
opener garnered 775! The ace lead produced
only a little over half as many marches and more than twice as many euchres.
Champ got lucky.”
I think so. But don’t count him
out. They don’t call him
“Champ” for nothing.
Natty Bumppo, author, The Columbus Book of
Euchre
Borf Books http://www.borfents.com
Box 413
Brownsville KY 42210
(270) 597-2187
[copyright 2008]
[next] |