Tourney Bounty equal to buy in (1 Viewer)

[Mr Winberg]

To put it bluntly, I'd suggest that has more to do with the rest of the field's lack of skill, than it does with the winner's skill. The winner has to be good, of course, but if he's winning that much, my guess is that the rest of the field are either all novices, or at least they play like it.

Depending on the timeframe, there is a third option...

 

[Jonesey07]

I had considered the idea some time ago, of a once a year, all-bounty event. Before I did, I ran some numbers. I could go back and re-examine, but my group and your group may be very different. So I will instead summarize...

• Strong players that typically finish at the top would see a dramatic pay-cuts. Dramatic. Numerous times the winner would win $60 in a $20 tournament with 18 players. That's little money for 5 hours.
• Heads-up would be as boring as watching paint dry. It's basically a very flat payout at that point, and I suspect chops would be more prevalent. It's my belief that a chop is a sign that the game is not fun for the players anymore.
• A lot of poor players would have a better chance to break even, or even double up.
• Nitty players would fair very poorly. That guy that limps to a just in the money finish... yeah, he's losing every time. Discouraging one type of play makes the game more 1 dimensional, and thus, less skillful.

We should have done this on Saturday, IMO.

 

[BGinGA]

While agreeing that one should wait and see what the results are over a larger sample size for this player... I can’t agree that tournaments are more about luck than skill in the long run. Especially for smaller tourneys with a fairly consistent group of players.

I host a two table tournament twice a month. The game has been going on for about a decade; I am the fourth host. While there are a few casual participants, and we’ve added or lost some players, the core group has been remarkably steady.

For about six of those years, including the last three, I have kept careful record of all the results—how every player has placed. About half that time, it was a weekly game, so I have data on about 250 sessions.

These results have been kept in spreadsheets, allowing me to really analyze where people are placing, how often they’re getting in the money, who is a truly winning player and who is not.

It is very clear to me that among this pool of players (currently about 16 of us who play at least 2/3rds of the time, plus another 8 occasionally), there are clear winners and losers.

Generalizing a bit, there are roughly 1/4 whose results are noticeably better, and 1/4 who are noticeably worse, and half who fall in between.

This is not to say that the top 1/4 always are the top finishers. Everyone has bad nights and good nights. But some have a lot more good nights and a lot fewer bad ones.

There are also some on the edges of the top category and the average ones, who often place either just before or just after the bubble. These may be slightly winning or slightly losing overall... generally more nitty players who last longer than average but rarely go deep.

Then there are a couple outliers who tend to win it all a bit more than the average, but often place at the very bottom if they don’t go deep—these being the wilder players, who amass huge stacks a small percentage of the time by taking lots of chances and sometimes running hot.

But without question, in this group it is obvious that while anyone can win on any given night, if you had to bet for or against people, you’d know who to pick...

^This almost exactly echoes the examination of our poker tourney group's results over the past 16 seasons and roughly 400 events, with around 100 total paticipants and 16 core players over that time span. Of those core players, about four are considered well above-average strength, four are considered relatively weak players, and everybody else swims in the middle-ground waters (at various depths).

Over half of those 400 events were league events with scoring that was heavily performance-based (vs attendance-based), and those cumulative data results clearly demonstrate who's who in the skill hierarchy (most wins, most cashes, most points, most knockouts, most season titles, most championship game appearances, most championship titles, etc.). It's a pretty small amd elite group of repeating players, regardless if the results are viewed on a by-appearance basis or not.

Thank goodness @Ben doesn't play here.

 

[Poker Zombie]

We should have done this on Saturday, IMO.

This is what that game would have looked like, with buy-ins deducted...

Situation	Actual	Bounty Only
1st, aka the run-good deity	$110	+$220
2nd	$60	-$20
3rd	$40	-$20
4th	$10	+$20
Two players	-$20	0
3 players	-$40	-$40

I'll let the group guess which player you were

 

[Jonesey07]

This is what that game would have looked like, with buy-ins deducted...

Situation	Actual	Bounty Only
1st, aka the run-good deity	$110	+$220
2nd	$60	-$20
3rd	$40	-$20
4th	$10	+$20
Two players	-$20	0
3 players	-$40	-$40

I'll let the group guess which player you were

I'm trying to raise my fantasy poker draft stock.. haha

 

[CrazyEddie]

One way to measure how much of each player's results are due to luck vs. skill is to measure their skill based on their results! This is what game skill rating systems like Elo, Glicko, and TrueSkill do.

Glicko would be a good choice here (specifically Glicko2). Its model isn't designed with poker in mind, but you can make some simplifying assumptions that map onto the Glicko model without doing any grave injustice to the model's validity. Specifically: Consider each tournament as a round-robin pairwise tournament between all the players, where in each pairwise matchup the winner is whoever placed higher in the actual tournament. This is convenient, actually, because the Glicko model needs a) a rating period, which is a time within which all matches are assumed to have occurred more-or-less simultaneously, and b) ten to fifteen matches during that rating period, with five to ten matches per player. So one poker tournament maps nicely to one Glicko rating period.

A nice feature of Glicko unlike Elo is that you get not only a skill rating but a confidence interval for that skill rating. So whereas Elo might tell you that someone's rating is 1600, with Glicko you learn that their rating is 1600 but that their actual skill is very probably (with 95% confidence) somewhere between (for example) 1500 and 1700.

If you run a Glicko analysis of all of your players and tournaments and get results showing that their ratings are far apart and their confidence intervals are small, then you know that their results are largely due to skill. If the analysis instead shows that their ratings are close together and their confidence intervals are small, then you know that the results are largely due to luck (here meaning not that they aren't skilled, but that they are roughly equally skilled). And if the analysis shows that their confidence intervals are large, then you know that the results might be due to skill, but they also might be due to luck, and you just don't have enough information to know for sure. Yet.

Here's an Excel plugin that will calculate Glicko ratings for you. I haven't tried it. I don't know if it will show the confidence intervals as well as the ratings; hopefully it would. https://www.add-ins.com/free-products/chess-ranking-assistant.htm

 

[Blind Joe]

To put it bluntly, I'd suggest that has more to do with the rest of the field's lack of skill, than it does with the winner's skill. The winner has to be good, of course, but if he's winning that much, my guess is that the rest of the field are either all novices, or at least they play like it.

Can't argue with this. I don't have a huge pool of players and I'd say probably only half really know more than the basics, some are brand new because I'm always trying to build my group, so yeah, he isn't playing against the toughest table.

I'd say of the regulars, 4 of us are pretty competent (don't get me wrong we would almost certainly be fish in most of your games), but it's usually that group which makes it to the later levels, makes the money and we've all won tournaments. Modesty aside I would have considered myself to have a slight edge on the others and the stats seemed to back that view up.

The new top dog is good though, there's no taking that away from him. While the rest of us "good" players have enough skill to beat the fish he is on the next level that we need to get to. He's got a good poker brain, he's good at working out the range of familiar players, he'll sit there and work backwards through the action and has on more than one occasion placed me on my exact hand. That may say more about my own style and game than his reading ability but still, he's good.

 

[stavman]

This could be fun, might try this at one for Tournaments.