Currently trying to figure out how to work out how the bounties should affect our shoving ranges. Some -EV shoves now become +EV when we cover our opponent due to the additional KO equity and calling all-ins wider now makes sense. Can't find anyway to factor in the Bounties since Holdem Resources Calculator and ICMIZER2 don't have the function built in.

Say we have a standard bubble spot blinds 50/100-10 and stack distribution
BTN 900
SB 1000
BB 1100
Regular nash with the 50/50 payout structure looks like this


But given the $EV of calling hands like KQo and some Axo hands will now become +EV due to the bounty how do we figure out what the new nash ranges are given that BB plays taking the bounty into account? How wide will small blind be able to shove profitably now?

As a first-order adjustment, add twice the bounty amount to the first place prize, as the winner automatically earns the bounties on himself and the runner-up.

As for the immediate bounty that can be earned by knocking SB out (which is $12 in a $25 KO hyper, whereas the 2nd prize is $36 and the 1st prize with 2 bounties is $60, and $12 is 12.5% of the effective regular prize pool of $36+$60=$96), take into account that even if Hero folds, there's still some chance for Hero to knock out someone in 3rd.

If Hero's chip lead were big enough not to be affected by forfeiting 1 bb, I'd say that he'd be able to knock out an opponent in 3rd in a further hand at least half the time as per the Malmuth-Weitzman model, and there would be no hurry to stack off in this very hand, but the case that you've posted is particularly difficult and must involve heavy FGS, as SB takes the chip lead if Hero folds, and if Hero takes most of SB's stack in a further hand, there will then be pretty much a lottery between Hero and BTN for the remaining stub of SB's stack.

If we estimate that Hero's chance of taking the 3rd place finisher's scalp drops from ~1/2 to ~3/8 as a result of the effect from the previous sentence, then Hero essentially passes up on 5/8 of SB's bounty if Hero folds, thus I'd make a call with a negative edge down to (5/8)*12.5% ~7.7% of the prize pool, as calculated using the effective prize structure of $36/$60 for the 2nd/1st, but bear in mind that, because Hero has to call wider, SB has to exclude a lot of semibluff hands from his shoving range and weight his range towards thin value shoves (face-rag hands), so you need to adjust SB's range manually.

Thanks for the response. Interesting way to treat the payout structure but it makes sense. I think that the future game aspect is definitely the toughest part of these games so far.
Going to be interesting trying to figure out correct strategy in these weird spots. But what you posted is a great start.

You might also want to send Q (from ICMIZER) a note on how to use the tool to adapt as well. He is always willing to provide suggestions.

I think the same applies to plexiq, the author of HRC.

All that said, I doubt they'll ever want to add a full support of bounties - the bounties make computations much more complex. A really good tool for 6-man SKOs would cost 4 figures and require a full night to process a spot like 6-max cash solvers. As for calculating the 18-man bubble strategy (with variable-depth FGS because a fixed depth of 4 hands doesn't quite do the job) - forget about it, I doubt modern hardware can handle it within a week even with optimisation.

Haven't done any work on these yet (and am unlikely to, tbh) but I love how complicated the bounties make it to 'solve' the game and hope Stars implements similar tactics in the future.

just go all in first hand. If you take the blinds, you will have the opportunity to bust every player but no one can bust you in 1 hand. This means they will have to call tighter as they risk their stack only for chips and not for KOs. In reverse you can "wide way shover" (this is a typo but it's so messed up, I let it stay) errr shove wider because they tighten up and you increase the difference in stacks in any hand. If they call, well you still can bust them and get the bounty - also with the benefit of no one busting you. If you lose the race, reg another one.

cliffs: KOs are the best game for a drooler to play 100/100

Ditto.

What do you guys think about variance in those games?

Due to a flatter distribution of prizes (one doesn't have to get HU to cash, earning a bounty does the trick too), it's even less than the one of usual 6-man hypers, but more than the variance of DONs of other sites.

I like the format, but the rake is pretty absurd in the 18 and 24 man games.

You make it sound like 6man hypers are low variance Thanks for your imput though

Poker is never easy

They are (when there's a fish at the table at all) - in comparison with Spin & Gos and especially MTTs; I didn't want to compare them with dead, solved or otherwise beginner-unfriendly formats.

So far seems to be a really rec-heavy/friendly. Lot of the regs seem to have no clue what they're doing. Any idea how much of our winnings will be coming from Bounties vs Placing 1st/2nd. I assume it would change based on strategy of course.

Had some fun four-way all-ins pre.

The bounty is (ldo) weighted chipev along inflection points. Easier said than done to be truly exact. How significant the weight is inverse to chipstacks, so the only time the weight is relevant in fgs where most of the advantage is already accounted for by the end points. Therefore, chipev-bounty icm fgs is fine.*
*fine as it is in 'regular' icm games
jmo

My strategy is bounty-heavy like Uhrenknecht's, but in the 412 test 18-man games, I've only earned 482 bounties (1.17 per game) with fairly neutral luck, which hasn't been a sufficient compensation for the inherent lack of podium prizes (62, i.e. 15% ITM), so I must be doing something wrong

Plugged some numbers into my excel to try and figure out how much I'm getting in KOs. seem to be winning slightly more from placing than KOs. But yeah lol sample (220) + crazy ITM%

is ITM = In the money or ITM or = I won more money (KOs combined) than the BI is?

I meant the frequency of getting a prize for placing - KOs excluded.

Do the higher buyins help to make more money on KOs?

It depends on one's skill level relatively to the field. I normally avoid the highest buy-in on offer in any new format (Spin & Gos or newer) because that's where the overskilled elite 'gangs up'.

I'll make an exception and try to grind the $5 Duel from the start, though, in hope that there will be few regs even in the 'challenge a random player' mode, because the $2 one won't be promising in terms of hourly profit anyway.

I am wondering when 6max KO hyper games were introduced? I think that when Neymar showed in the KO tab and with that PS introduce a lot of new ko games.

Very funny games are those 6max KO hypers Like it a lot.

What do u think is good roi at 3-5$ level games pre rb?

I assume at $3-$5 the games will be pretty soft. So far I have a 700 game sample mixing at the $25 and $10 level. A bit of game selection here and then (just avoiding games with 4 other regs) but run at 12% ROI so far. I think anything above 5% is considered really good for hypers. The higher rake at micros definitely reduces the ROI by a bit but that's off-set by softer games.

Hi!
I played about 1k 6max ko micros to try them and saw the field is so weak (a lot of fish and tight -in my opinion 22/18 is very tight for this game- regular players). I should assert that I'm able to calculate exact bubble factor in particular cases only (no FGS included) and I'm in line with previus post about the possibility of ICM softwares of incuding this structure definitively.
What do you think about bb/100? Do you think 10-11 ev bb/100 overall (in a first approx obv) are decent in order to get a >5% roi in 15-20k sit?

Hi guys.
Full support for bounties in KO tournaments was added to ICMIZER 2.
It uses a tricky model, but it works instantly. FGS support is planned for near future too (currently KOs do not allow FGS on).