Can anybody give me a quick answer to what varience is in MTT/online tourneys? Trying to get my head around a lot of things I need to study for in depth. Thanks

(nooby question I know lol)

Variance is a poker players code word for 'luck', and usually refers to when the actual outcome varies from the expected outcome.

Outcomes are shaped by the behavior of the participants, so game formats that feature or promote non-standard behavior are higher in variance.

Consider two MTT's. The first is an online freeroll. There is no cost to play, no time commitment or expense in getting to the tournament, and there will be another one running as soon as this one ends. The other tournament is the WSOP main event, a huge event that costs $10K, requires prep, travel, and other expenses, and happens once a year.

Now consider this situation, in the early portions of a tournament, you wake up with KK and three bet to 10x. This action should narrow the field and improve your chances of winning the hand. In the online free roll, however, every limper and the opener call you, and you go to the flop facing 5 players. At this point, you aren't just afraid of the random ace, there are a lot of two pair combos or ugly low straights that can beat you as well.

In the WSOP event, however, your bet increases to pot size and signifies strength, so players without premium hands will tend much more strongly to fold, and you end up isolating with the original opener. Your chances of winning are much higher against one player. So variance, the liklihood that unlikely outcomes will occur, is reduced

A lot of factors influence variance, usually they are related to the cost of busting out. Is the touranament expensive? If you bust out, are you done for the day or can you rebuy? Have you committed a lot of time to the tournament? Is there another tournament starting right away? Other factors, like blind structres, can also effect variance. If the blinds are increasing rapidly, players have to play a wider range of hands, as they will very quickly be blinded out if they wait for premium hands only.

I think the easiest way to illustrate variance is the winner take all format. It takes a bigger sample to realize one's true winrate in the winner take all format. Compare that to a satellite that pays a large portion of the field. Clearly, your actual winrate will be more likely to converge to your true winrate faster in the satellite.

I wouldn't play poker if it didn't have variance.
I love sucking out in MTTs and have people abuse me in the chat while im quiet.
I also love making my opponents make mistakes. if i win or lose the hand its great.

without variance in poker it would suck.

you need that mindset otherwise you will quit when you lose. and you will lose you cant win all the time. that goes for any competitive sport as well.

Contract Bridge is also a beautiful game.

The expected ROI of the players simulated players in this graph is 10%, with an EV of $50. As you can see one luckbox wins about $700 and one runs horribly and loses nearly $300. 40% of the simulated 10% ROI players will lose money in their next 500 tournaments.



Put your own numbers in the calculator at http://pokerdope.com/tournament-variance-calculator/

Actually, variance isn't anything you need to "study." It just happens to you. There are so many other things you should be spending your time working.

There are two different definitions of variance. One is the mathematical definition, which recognizes that the results will vary across a mean result. The other is the poker definition, which only occurs when you are losing. If you are winning, it is due to skill. If you are losing, it is due to variance.

In Arty's example, the guy who won $700 almost certainly believes he is a crusher in the game. He'd laugh in your face if you told him he was no better than the guy who lost $300.

That explains it well for me thank you so in bigger tourneys with bigger buy ins maybe like WCOOP varience is lowered as peope don't want to call with anything crap as appose to a 5 dollar tourney with loads of rebuys

Thanks for the replies everyone helped me understand it a lot better much appreciated

No, variance is not lowered. Variance is 'measured' after the fact. A player's 'range' may be smaller in a 'more serious' event but that doesn't necessarily change their variance. They may play less marginal spots/draws but a variance calculation needs multiple data points. The more data, the more reliably you can analyze the results.

Play 5 exact hands (almost impossible to do in poker) ... 4 worth $10 and the other worth $100. You are 80% to win but only win 3 (60%). There are only 2 possible results.

1) Best case you win $120 in chips ... you 'invested' $70 (5+5+5+5+50)
2) Worst case you win $30 in chips ... you still 'invested $70

Who had better variance? Neither, both lost 2 hands which is 20% below the expected result (80%) for that spot. Where the variance discussion gets muddied is when one of these players starts talking about 'playing' better than the other based on him winning $50 v losing $40. Both players 'played' the same by putting their opponent in a losing spot, 80-20, five times in a row. But one of the players 'ran better' by winning the big hand.

Being able to separate your play from your results is very important to long term poker success. The increasing number of tools within poker software can help you 'see' that even though you may be down in your BR that you are making the correct plays but just falling below the expectation.


Start side story:

I just posted the other day about tanking in a potential 8.5 Buy-In pot against 2 opponents. I knew/figured I was at least a 55% favorite but didn't know if I wanted to risk the variance of losing a flip for 'all' my profit of the day. It was late in the session and this one hand would erase the previous 'hard work' at the table that day.

You aren't supposed to think this way when playing poker. If you are +EV you are supposed to just go with it ... Ultimately I did call and won the pot against an OESFD and overpair with middle set. I've played enough poker that I would've been 'fine' mentally if I had folded and saw the result. The real issue is that had I been stuck for the session I would've called without thought but instead I was trying to protect a 'win'. How you handle the emotional side of your game is as important as how you look at the stats.

Long story to show you that I was more than willing to risk 'variance' had I been down in the session ... and possibly leave an even bigger loser ... as opposed to 'preventing' variance from ruining a positive session with the same result .. losing the 'one' hand. GL

Indeed. All players are 'fooled by randomness' to some extent. I've got a buddy who will humbly admit he's not the best in the world (he only plays on a casual basis) but he's been running like an absolute god in MTTs this year, and it must be so hard for him to not think he's an all-time crusher.
A couple of days ago, he came 7th in the half-price Milly, and the stats for his <400 tourneys are otherworldly.



Fun fact: When he won a turbo deepstack earlier this year, he was printing money at a rate of 50 dollars per minute... for four hours. #MBN

Variance in big field MTTs is just completely crazy. On that graph, my friend has an epic 168% ROI (he won 258 buy-ins on Sunday night alone), but notice that he had a downswing of 150 tourneys (about $6500) after the first bink. Without his three big scores, he'd be a losing player.

Great example.. starting to build a lot more info through this site and forum where can I get graphs/charts to input my buy ins and results to see my ROI?

To track your results, you need database software (which has the added benefit of a HUD). The most popular trackers are Holdem Manager and PokerTracker, as these automatically import your hand histories and results, and they do all the calculations for you, and can produce all kinds of interesting reports and graphs.
You can get by with an Excel spreadsheet if you just manually enter your buyins and results and know how the graphing functions work. (I track my results on Unibet in this way).

Yeah I'm definitely going to start tracking my results using excel, as the manager programs, eventhough they come with HUDS, are expensive for me atm. What chart do you use for your unibet results on excel?

I just type in all the standard stuff (buy-in, number of players, finish position, winnings etc, like you can see in the pokertracker image I posted above) and I use a new sheet each week. At the end of each week I add the results to my 'yearly results' spreadsheet, so I can keep track of how many tourneys I've played in total, and use the program to calculate ROI and to produce graphs.

The top of the weekly chart looks like this:


The bottom of the chart is for the totals and averages that I carry over to the yearly chart. It's not really worth showing the numbers or the graph when I have a 0% ITM and a minus 100% ROI over this sample size though.

Very nice, gonna do a excel spreadsheet just like your one now thanks