POTD #290 Almost FT Friday: How Thin Can I Value Bet Near the Bubble?
A retirement gift for Fedor Holz
One of the most common questions I get asked nowadays is a variation of “At what point in a tournament does ICM start to matter?” Me being me, I struggle at giving a direct answer to a question, but I have been asked the question enough times that I’ve parsed down my long answer enough for it to (hopefully) make some sense. The short answer is, it matters from hand one, but the way it matters is extremely marginal– In a freezeout it folds to the BB who goes all-in for 100bbs and shows you 2d2s, do you call with Tc9d? I think most people understand in a soft freezeout, you might rather stay in the tournament than collect your 3% edge. They’d never fold in a cash game, which means your decision is being guided by, if not ICM, explicitly an idea that staying in the tournament is worth more than winning as many chips as possible.
So it’s always something you should be considering, but often the effects are very small. The metaphor I often use is that ICM affects your strategy in the way that wind might affect how you steer a boat. From hand one of the tournament, you are in the water and you want to make it to your destination; you cannot return to the port you came from. Throughout the journey the wind will shift in intensity and direction, but it’s always there. You need to adjust to it and hope you gauged the effects correctly.
Today’s hand is another tricky hand where I need to determine what my value betting thresholds should be as we’re approaching the money. I reverted to a simple heuristic of, just check back your lowest-EV value bets, which is usually appropriate, but in this hand I failed to realize what my lowest-EV value bet actually is, and I didn’t realize there were certain peculiarities about this hand that made it very unlikely my opponent, Fedor Holz, had me beat.
Triton Cyprus 2023 - Event #11 $100,000 NLH - Main Event
(15k/30k/30k) (SB/BB/BBA) 23 Left. 15 Cash. Average is 1.1M
It folds to me (1.27M) on the button with T♥️T♣️, I make it 65k, it folds to Fedor Holz in the BB (915k) who calls.
Flop (175k) 8♦️7♦️5♣️: Fedor checks, I bet 115k, Fedor calls
Turn (405k) A♥️: Fedor checks, I bet 200k, Fedor calls
River (805k) A♠️: Fedor checks, he has 535k back, I check and and beat K♥️6♥️
What I Was Thinking
We’ve had a stretch of strong preflop hands in POTD. Yes, you raise TT on the button. On these sorts of flops with three low cards and lots of straight draws possible, you often pick a big c-bet size with your weakest and most vulnerable overpairs. This is a board where 99 might get cute and bet small or AA might get cute and check, but TT has no draw and is vulnerable to overcards. I played it as a pure bet and a large one.
The turn A is not what I wanted to see. This is not like raising UTG, where I might frequently bet a hand like A3s or AdTc as a bluff on the flop. I shouldn’t have all that much Ax in my flop c-betting range. However, Fedor should fold a lot of ace high on the flop and be aggressive with a lot of Ax preflop, so if TT was the best hand on the flop, it’s likely still best now and I can continue value betting my flopped over pair. The river A is good for me; it counterfeits whatever low-frequency flopped two pair Fedor might have, but more importantly, it means he has hands like A9 or A6 much less often. If TT was the best hand on the flop and the best hand on the turn, it’s the best hand on the river. Shoving for value is tempting, especially vs. someone like Fedor who could easily look down at a hand like Kd5d and decide to call off with it. However, if I checked and lost, I’d be just below average, while we are still only eight away from the money. It felt like it was more important to save a big chunk of my final 800k chips than to go for thin value, even if I thought my hand was usually best.
What I Got Wrong
For chips, Fedor occasionally but rarely leads the flop. I pure c-bet my hand and mostly pick a big size, I pure two-barrel the turn and mostly pick half pot, and I pure shove the river and checking loses almost 5bbs. But this is not a cEV hand. I have a slightly above-average stack, and two-thirds of the field will get a $175k mincash. Preserving my stack is important, and neither Fedor or I are playing cEV poker. So what does a postflop ICM output say for this hand? Fedor occasionally, but rarely leads the flop. I pure c-bet my hand and have a 50/50 split between a big size and a small size. I am about 50/50 between betting half pot and checking on the turn and … I pure shove the river.
So I think my preflop, flop and turn mix are all fine; They’re frequent plays in cEV and tournament scenarios. My river play is not. My hand has 84% equity on the river (compared to 89% in cEV), and Fedor can call me down with any unimproved flopped pair. Fedor is definitely capable of making big calldowns and I should have shoved. It is my lowest-EV value shove— 99 is technically a worse hand, but Fedor never has TT, 99 blocks slowplayed combos of 96, and A9 is more likely to peel the flop than AT. I thought I needed to tighten up my value-shoving range on the river, but this is not a spot to do so, in part because Fedor’s range has so much 8x/7x/5x that anything that beats an 8 is good enough to shove, even as we approach the money bubble.
Types of Error
Nitty river check back
Grade
This is a pretty specific river card. The solver doesn’t want to shove an offsuit two river, because then Fedor will have enough top pair or slowplayed two pair that TT is too thin a value shove. The ace river gives him fewer Ax and counterfeits his two-pair hands, and TT is upgraded into being a hand good enough to shove. The river strategy here, both value betting and bluffing, is pretty simple. If you beat a pair of eights, whether you have 99, KK, or A2, you shove. If you can’t beat a pair of fives— even if you have Ks9s, which has 12% equity— you have to consider shoving. I had an oversimplified strategy of “value bet tighter at the terminal node because of ICM,” without thinking about Fedor’s range enough. I should be value betting tighter, but an 84% equity hand with two-thirds pot to play does not cross that threshold.
C+