May 28, 2023

*** This hand was suggested by joekayak
0*-0  ?
61%
32%
4%
0%
0%
0%
Total votes: 219
joekayakI seem to run into this hand a lot. I'd like the computer to tell me....Try for 12 via 3-8 discard or try for 12 via 10-4 discard?
james500
3921 votes

Joined: June 2013
 
 
 
Sunday 3:09 AM
I usually choose 3-8 from this, and 2-9 from the similar arrangement A2449X. I'd also like to learn if I've been doing it right.
MiketheExpert says: Hi james. (2-9) tips even more in favour of the this discard I believe when holding the arrangement (A 2 4 4 9 X).
Gougie00
5728 votes

Joined: March 2008
 
 
 
Sunday 4:03 AM
Not crazy about a A-10 toss to my crib. I tossed 3-8 instead and I may be regretting it for the rest of the game.
JQT says: Having seen your comments for about fifteen years now, I would assert that you have intuitively made the correct decision, and that you would do so again, in spite of the momentary sorrow due to the poor Cut Card. This feeling of regret or "buyer's remorse" will usually only last the duration of the current deal, or at most, as you suggest, throughout this game. But that is the regret that must occasionally be endured by the stronger player, one who accepts that the correct decision won't always be rewarded instantly in Life or in Cribbage. The important thing is to continue to make the correct decision, regardless of a single bad outcome: because in the long run, you will be rewarded. I think if you examine WHY you decided to Toss (3 8) today, you would be able to defend your decision and say that, while it may occasionally be disappointing, it's still the correct decision, and the one you'll make the next time you see these same cards. But the regret you expressed is very characteristic of those feelings that we must endure and overcome when we "play the numbers" and avoid playing hunches.
Granny says: Very sermonly and appropriate for a Sunday JQT:-)
JQT says: There are those who state that psychology does not belong on the Cribbage Board or on the Chessboard, but this is tantamount to saying that restrooms don't belong on airplanes! After all, we are human, and so we may attach regret to both mistakes as well as to correct decisions that don't pan out as we had hoped. A computer can determine what it considers to be the correct answer, and make it every time, and that's why a human who is good at counting cards will be quietly ushered away from the casino and invited to never come back! Most humans are unable to shut off all emotions, and even fewer have the ability to keep count of multiple decks of cards, and so in order to perform at a high level, it's necessary for us to maintain a close watch over our emotions. Good players can also project a powerful presence, especially 'over the board' and in a "live" game, but I've seen it done online as well. This occurs especially when a strong player senses any hint of weakness in his or her Opponent, and can often employ sub-optimal tactics that will "rattle" many players. This is when the losing player will often shake his or her head, or in some cases, believe that this steamroller of an Opponent is cheating! We are human after all, and psychological forces are always at work and at play within us, but the skillful player will have learned to corral and harness these to a great extent, and it shall be reflected in their results. - j q t -
horus93 says: I find five leads, especially on the first hand, are good for rattling people, because the initial impression many people get is that you just learned the game yesterday, but when it pays off, and you make other good pegging plays, it gets in their head a bit at least sometimes
JQT says: Exactly. Against the typical "Kitchen Table" player, a 5 Card Lead that WORKS will 'rattle' them a bit, and even one that DOESN'T WORK will sometimes "rattle" them even more (they will under-estimate your ability, which helps you, not them). But against a stronger player, try a scheme such as avoiding the automatic and seemingly harmless PAIRING during a random play, even when you feel it's the better response. If and when you trick your Opponents, and thus avoid them from obtaining PAIRS Royal, only to later PAIR that now 'Lone Card' they invariably will still be holding, that card they think is now SAFE, it will stun many intermediate, and even some very strong players. (From an expert player, you might even draw a rare smile or nod! Consider this your only yet sufficient reward.) And, the beauty of such a tactic is that when you cannot make it work, which will be a fair percentage of the time, it will invariably go unnoticed by most Opponents. This is key to such tactics, as you can often use them over and over, sometimes even during the same game.
Gougie00 says: The game would be easier if I knew what the starter was.
Tigger23505 says: Yes, knowing the starter would be nice, but that's not cribbage.
HfxKen
885 votes

Joined: February 2020
 
 
 
Sunday 4:42 AM
HfxKen
885 votes

Joined: February 2020
 
 
 
Sunday 4:44 AM
HfxKen
885 votes

Joined: February 2020
 
 
 
Sunday 4:46 AM
Ok, I'll try this again. If I toss 10-Ace, 2 of the cards I'm hoping to catch in the crib or cut, are already in my hand.
horus93
1281 votes

Joined: December 2017
 
 
 
Sunday 4:53 AM
At 0*-0 with cards like this, am aiming at overall scoring potential in the discard. A choice of offense vs defense vs balance would depend on the cut.

3448 has an 11, but the 34 five trap is more apparent than real here, and AT to own crib is pretty bad.

A44T has a double nickle and an eleven of its own. As Ras pointed out in one of his classes, if you look at the deck in a certain way, about 40% of it is nickles and dimes - or, at least, nickles, dimes, and potential nickles - so this sort of configuration is powerful. 38 isn’t a great crib toss, but it’s a bit better than AT anyway.

I feel like we had this same hand a few weeks ago? Or maybe it was just something similar, or it was for pone or something.

Per Liam this keep is on top at 12.6 in hand+crib. 3448 is in second at 11.6 in hand+crib. I don’t think the pegging between these two hands would make up that gap, although it might end up kinda close if choosing an offensive strategy because 3448 could create runs and whatnot against some pone hands. But we shouldn’t ignore defensive pegging at this score, and on that front, I suppose A44T is a bit better, for the same reason that 3448 would be a bit better on offense.

The cut hits the three in the crib, so there is room for hope there. Still, with only 8 known points (6 for the hand and at least two in the crib) + one guaranteed peg, would try to get some points in the play, although I wouldn’t pair an A or T lead.
horus93 says: Obligatory typo: this is 12.2, not 12.6. If it’d been 12.6, would not think the gap would narrow much even on offense, but at 12.2 it likely does at least somewhat
MiketheExpert
1120 votes

Joined: April 2021
 
 
 
Sunday 5:40 AM
This is actually a pretty close decision. (A 4 4 10) has an ever-so-slightly higher expected average because of the additional points on face-card cuts, however (3 4 4 8) has a higher maximum (14 pts) and more cuts for a dozen+. Both throws to own crib have negative delta surrounding them holding the 4's (A 10), and (3 8). (3 8) is still a superior throw, and going on general principle of defense as FHD, (A 4 4 10) is also safer in terms of likely pone pegs. So I'll hang to (A 4 4 10) usually on the first hand. If I had known the cut would be a deuce in advance, of course I would have changed my mind.
MiketheExpert says: Still the deuce could work with the (3 8) in the crib, but unlikely to make up for the 6 missed points in hand at least.
horus93 says: Very good point on means vs distribution. I agree that if you think exclusively in terms of means, like the bot, you’re going to make some miscalculations.
MiketheExpert says: In terms of pure calculations, HAL should reflect the superiority of the (A 10) discard in all 3 facets (i.e. Hand average, Crib average, and pone pegs against). This should make the overall difference in the neighborhood I am guessing of 0.7 - 0.8 points. Notwithstanding this, There are still places on the board and strategies where I will go with for the 16 cuts for 12+, in favor of this, especially if desperation offense may be required.
MiketheExpert says: Excuse me...not (A 10), I meant (3 8)!
JQT
4143 votes

Joined: October 2008
 
 
 
Sunday 5:56 AM
Today, joekayak gives us a card arrangement everyone has seen, but many of us have debated over which Dealer Discard is correct. He asks us which path to a Dozen Points is best. Let's see if we can finally put this familiar puzzle to rest.

After Keep (A 4 4 T) and Toss (3 8), Twenty Cuts (AAA, 44, TTT, JJJJ, QQQQ, KKKK) or 20/46 = 43% yield Ten Points or more, and Eight Cuts (AAA, 44, TTT) or 17% yield a Dozen Points, however, Fifteen Cuts (2222, 3333, 888, 9999) or nearly 33% add NO HELP whatsoever.

After Keep (3 4 4 8) and Toss (A T), Twenty Cuts (2222, 333, 44, 5555, 7777, 888) or 20/46 = 43% yield Ten Points or more, and yet Sixteen Cuts (2222, 333, 44, 5555, 888) or nearly 35% still yield a Dozen Points or more, however, a whopping Twenty-Five Cuts (AAA, 6666, 9999, TTT, JJJJ, QQQQ, KKKK) or greater than 54% add NO HELP whatsoever.

Being able to get those additional Two Points TWICE as often, or even the ability to get Fourteen Points about 4% of the time after a 4 Card Cut, will NOT overcome the fact that Keep (3 4 4 8) fails to get any help greater than HALF of the time. In addition to this, there is at least a HALF-POINT difference in the Dealer Discard between Toss (3 8) and Toss (A T).

As we can see, this puzzle involves a bit more than merely asking about those Dozen Points; the solution actually involves how often we shall begin with just Six Points! Therefore, having done the work, we'll Toss (3 8) and retain the slightly more powerful Hand today, with the slightly better discard.

After the Deuce Cut, we still have just Six Points in our Hand, but with prospects for a good Crib. This puzzle is an excellent illustration of my frequent observation that Cribbage is indeed NOT a self-correcting game! Simply playing more games can only bring your level of play so high: at some level, you MUST study the intricacies of the numbers to become a stronger player!

For those who decided to Toss (A T), after the Deuce Cut, you shall have reached a Dozen Points in Hand, yet with prospects for a bit lethargic Crib. But if a player does Toss (A T) and gets rewarded for it, now it may be months, years, or a lifetime of thinking that an inferior discard is superior.

Wordle 708 4/6 (my favorite verb, but only when it describes a bikini 👙)

🟩⬛⬛⬛⬛
⬛⬛⬛🟨⬛
⬛⬛🟨🟨⬛
🟩🟩🟩🟩🟩
Eolus619 says: JQT..as you have pointed out many times..cribbage is NOT a self correcting game due to the random nature of the cut !…and I get nervous when it takes you four guesses on wordle,.See how this goes for me then
Eolus619 says: oh…👌on your clue…
Eolus619 says: JQT…took me five …these seldom used letters are troublesome to me!
JQT says: Errata: I counted one extra Trey above in the Cut Cards that DO NOT add any help for (A 4 4 T), thus only strengthening my argument even further; and therefore, the last line of the Second Paragraph should read: Fourteen Cuts (2222, 333, 888, 9999) or about 30% add NO HELP whatsoever.
JQT says: Five Guesses in Wordle still keeps your STREAK intact! In my current stats, my percentages for each number of guesses is as follows: 1 (0.00%), 2 (2.201%), 3 (22.956%), 4 (44.654%), 5 (22.327%), 6 (7.862%). Therefore, I have over 22% Guess Five solutions. Most importantly however is that my streak has continued for nearly a full year! There are many ways to "play the game" and by definition, winning is simply getting the solution within the allotted Six Guesses. Some players strive for the best or lowest average score (mine is currently 4.044), and there is a way to calculate this mathematically. My own stats involve no use of programs and no cheating whatsoever. I have read that The New York Times has found that as a nation, Sweden has the best or lowest average score (about 3.72), and that North Dakota has the best score (about 3.65) among the United States. I believe the country and state results here are accurate, but I suspect that all such aggregate numbers likely include some players who either cheat, get help, or use various tools to make solving the puzzle much easier. Ref: https://whatisthisthing.com/wordle-average-score-calculator/ and https://www.reddit.com/r/wordle/comments/tcvwb8/how_do_i_calculate_my_average_score/
Eolus619
1340 votes

Joined: June 2020
 
 
 
Sunday 5:56 AM
Interesting puzzle containing many cribbage axioms 1) GFD needs eight points to maintain advantage 2 ) the crib is an extension of the hand 3) Tas tells us to look for “nickels & dimes” 4) the most important card in cribbage is the cut & it is random.
Eolus619 says: actually Tas is Ras in his disguise..whew
JQT says: What surprises me most about today's puzzle is that, as of nearly 10a ET, none of the comments have said that the real reason (A 4 4 T) is better than (3 4 4 8) as First Dealer is not due to the number of Cut Cards that ADD points, but because of the fact that it has fewer Cut Cuts that DO NOT ADD POINTS! Regarding your postscript: If "Ras" has suddenly become "Tas" after just two upgrades, think of how he might feel after only three more promotions, when he shall become past-tense or "Was"! P.P.S.: How did you do on that Wordle today?
horus93 says: Isn’t it the same thing, to have fewer cuts that don’t add points, and more that do?
JQT says: That's what makes this such a great puzzle, because both Hands have the same number of Cut Cards that yield Ten Points, but (3 4 4 8) actually has twice as many Cut Cards that yield a Dozen or more Points! And yet (A 4 4 T) is better because it has many FEWER losing Cut Cards! So the answer is NO: Having more Cut Cards that add points is not always the inverse of also having fewer Cut Cards that do NOT add points! And alas, having more Cut Cards that FAIL TO HELP can sometimes trump a Hand that actually has more Cut Cards that add the most help, as is the case today. Read my earlier post (along with the errata) for the actual numbers and percentages.
horus93 says: I see what you mean now. In terms of number of helpful cuts vs having fewer that miss, this is of course the same number, but you were referring to the distributions and how those work out
Inushtuk1 says: But if we were Pone needing 14 to go out, I would imagine 3-4-4-8 would be best.
horus93 says: Yeah, and I’d say at a score like 65*-77, where we need not just an overall good pick, but a high hand b/w fireworks in the pegging to have a decent shot at winning, 3348 would probably be best
horus93 says: Errr 3448
hecklebush
203 votes

Joined: February 2022
 
 
 
Sunday 6:35 AM
I always want to play my crib, but sometimes the hand dictates the discard. Holding the 344 is a nice outside straight, and it paid off in this case with the 2 cut for 12 points.
Inushtuk1
1485 votes

Joined: July 2016
 
 
 
Sunday 7:02 AM
I see the error of my ways. I got lucky. Next time I will only play it this way as Pone. Defense to the lead.
Inushtuk1 says: Actually A-4-4-10 is best for Pone as well. See May 8th of this year.
J.W.B.
353 votes

Joined: March 2014
 
 
 
Sunday 7:16 AM
Eolus619 says: welcome..haven’t seen u post before ..keep doing so!
Eolus619 says: don’t forget to add your thoughts/comments/logic/reasons
GT25Ump
285 votes

Joined: August 2016
 
 
 
Sunday 8:05 AM
I so wanted to put an A-4 in the crib, but it was just going to really downgrade my hand and it'd just be my luck that no facecards would pop up to help out that 5, ha. So, kept the sure fire six points with a chance of a run...and oh, looky what card gets cut, nice start out of the gate.
winesteward48
833 votes

Joined: April 2021
 
 
 
Sunday 8:56 AM
In my mind it was a close decision to keep these cards or A-4-4-10. This time the more chances of 12 points won versus no improvement. Next time, depending on what I eat for breakfast, tossing 3-8 might win.
Ras2829
5152 votes

Joined: November 2008
 
 
 
Sunday 10:23 AM
Choosing defense as opening dealer (matters not whether choosing offense) combined values give a significant edge to holding A-4-4-X. The slightly larger hand value and the much higher crib value of 3-8 (4.130) compared to A-10 (3.509). It appears that negative delta would be equal for either toss with 4-4 in hand. Think that computes to about .16 per card held. The margin in value would not change favoring the 3-8. Will avoid pegs except for those which seem reasonably safe.
Inushtuk1 says: Hi Ras. But wouldn't the better offensive pegging make up the difference for the Dealer if we were on Offense?
Ras2829 says: Hi Inushtuk1: The crib average is so horrid for A-10 as compared to 3-8, the small difference in pegging potential (about .25) is overwhelmed. The averages for A-10 and 3-8 are so different (rankings so profoundly different that if you arranged the 91 discards into three pages) with the higher value discards on page 1 and the lower average discards on page 3, 3-8 would be on page 1 and A-10 would be a page 3 discard. No, the better offensive pegging would not make up the difference.
Coeurdelion
5593 votes

Joined: October 2007
 
 
 
Sunday 10:41 AM
I think it muat be a straight choice between A-4-4-10 (3-8) and 3-4-4-8 (A-10):

A-4-4-10: 6pts + 4pts (Schell: 3.92) = 10pts

3-4-4-8: 6pts + 3½pts (Schell: 3.42) = 9½pts

Potential:

A-4-4-10: Improves with AAA, 44, 5555, 6666, 7777 + 15xXs = 32 cuts = 32/46 = 69.6% up to 10/12pts with AAA, 44 + 15xXs = 20 cuts.

3-4-4-8: Improves with 2222, 333, 44, 5555, 7777, 888 = 20 cuts = 20/46 = 43.5% up to 10/12/14pts with all cuts.

Position:

As First Dealer positional hole is at 8pts so I'll play Defense but try to score the average 16pts or more.

Pegging:

I think both hands will peg well but playing Defense perhaps A-4-4-10 slightly better.

Summary:

A-4-4-10 is better for starting value by ½pt and it also has more cuts for improvement although A-4-4-10 has 20 cuts for 10/12pts and 3-4-4-8 has 20 cuts for 10-14pts. Even so I'll go with A-4-4-10 for the better starting value and more cuts for improvement and throw 3-8.
HalscribCLX
5316 votes

Joined: February 2008
 
 
 
Sunday 10:44 AM
At 0*-0 playing a Defense strategy for the pegging the dynamic expected averages and Win/Loss %s are:

_______________Pone's
Defense___Hand_Pegs____Crib_Total____W9 %____W10 %
A-4-4-10___8.61+(-2.07)+3.67=10.21____36.1____49.5
3-4-4-8____8.52+(-2.04)+2.99= 9.47____35.2____48.4

Defense_______L9 %____L10 %
A-4-4-10_______26.0____24.7
3-4-4-8________26.2____25.6

A-4-4-10 is better for expected averages by 0.74pt. and is slightly better for Win %s and lower for Loss %s. So I'll select 3-8 to discard

After the 2 cut I'll play Defense to the lead.
dec
6356 votes

Joined: April 2008
 
 
 
Sunday 10:46 AM
Defense and more effective cuts. The means dictate this choice the other one is disregarding the face cards. Defense on the lead. dec
mike320
19 votes

Joined: April 2023
 
 
 
Sunday 1:02 PM
At the start of the game this is an easy decision. I don't see any reason for a long explanation.
hecklebush
203 votes

Joined: February 2022
 
 
 
Sunday 2:03 PM
Furthermore, I always say, "first hand never counts, last hand always counts..."
bsherms
284 votes

Joined: May 2022
 
 
 
Sunday 3:14 PM
I’m not sure why I chose this hand except that 38 in my crib felt better than AT. I can see I’m in good company!
JQT: is it cheating when I read your post before attempting Wordle just to see if there’s a clue?!
JQT says: Hopefully, my clues are rather obtuse (good word for six-letter Wordle!), as I would never intend to spoil a puzzle. I try to say something that will only make sense in hindsight, which is very tricky. Maybe Bruce can say if I've ever really helped. The few times I did leave better clues it was always very late in the day, like this comment at 8p ET.