How Hard is it to Move Up in Stakes?
Posted on Sep 14, 2009 by Gugel in Data Analysis
So as a follow-up to my previous post about data analysis, I decided to find out how hard it is to move up in stakes. Let’s say you’re playing $100NL and want to move up to $200NL. Just how much tougher is the competition?
Up to now, it’s been all hearsay. One guy says it’s easy. Another guy says it’s hard. There was just no way to objectively measure the difficulty in moving up in stakes. But now, thanks to PokerTableRatings, and lots of data analysis by yours truly, we can have a better idea of what to expect when we want to move up.
So here’s what I did. I assumed that the average number of hands played at a certain level is an accurate reflection of the competition. The more hands played, the weaker the competition. The less hands played, the stronger the competition. The data below supports that correlation.
| Stakes | Average HU Hands Played Per Day on FullTilt |
|---|---|
|
$50 NL
|
263,158
|
|
$100 NL
|
128,205
|
|
$200 NL
|
74,627
|
|
$400 NL
|
40,816
|
|
$600 NL
|
15,780
|
|
$1000 NL
|
15,152
|
|
$2000 NL
|
6,944
|
|
$5000 NL
|
4,878
|
Now, as you can see, there’s a big drop in the number of hands played when you go from $50NL to $100NL and relatively small drop from $600NL to $1000NL. So the theory is that:
- A big percentage decline in the number of hands played from the previous level means it’s hard to move up
- A small percentage decline in the number of hands played from the previous level means it’s easy to move up
So in other words, it’s relatively tougher to move up from $50NL to $100NL than it is to move up from $600NL to $1000NL. Here’s a pretty graph to illustrate that point. The lower the Difficulty Index, the easier it is to move up to that level from the previous level.
To make it even clearer, here’s an easy to read chart.
| Stakes | Difficulty of Moving Up from Previous Level 10.0 = Most Difficult 0.0 = Least Difficult |
|---|---|
| $50 NL | N/A |
| $100 NL | 5.0 |
| $200 NL | 4.0 |
| $400 NL | 4.5 |
| $600 NL | 6.0 |
| $1000 NL | 0.5 |
| $2000 NL | 5.5 |
| $5000 NL | 3.0 |
But I wasn’t done yet. I wanted to squeeze every drop of insight I could out the data. Maybe some levels have fewer hands played than what we’d expect. If that’s the case, that would mean that heads up poker “market” has not properly adjusted to the market conditions and there was a window of opportunity.
The blue line is the actual average number of heads up hands played on Full Tilt per day. The red line is the predicted number of hands played. As you can see, it looks like the HU poker “market” has adjusted pretty damn well. Damn the efficiency of those free markets!
- $100NL and $600NL are slightly undersaturated
- $1000NL is slightly oversaturated
There aren’t really any big windows of opportunity, but that doesn’t mean you shouldn’t move up. All you have to do is win more than half your current winrate to make it worthwhile! Just get ready for some sick swings and unless your moving up from $600NL to $1000NL, don’t expect it to be easy
Buy Hand Histories
The more you know about your competition, the more you can profitably exploit them. You can buy hand histories to boost your Hold'em Manager or PokerTracker database and help you better reads on your opponents.
| Site | Price | Coverage | Website Quality | Overall Rating |
|---|---|---|---|---|
| hand-histories.com | $$$ | |||
| hhdealer.com | $ | |||
| pokerhandscout.com | $ | |||
| PokerTableRatings.com | $$ | |||
| handhq.com | $$ |




anon
14. Sep, 2009
How are you determining “predicted” hands played?
Gugel
14. Sep, 2009
@anon
The “predicted” number of hands played is just that – a prediction. It’s a very rough estimate to see if there are any significant opportunities.
Anyway, the “predicted” hands is based on a constant percentage decline between stakes. So in other words, it assumes that the number of hands played should decrease by 52.5% as you move up in stakes.
If the red line was far below or above the blue line, it would be worth further investigation, but that wasn’t the case
dadada
11. May, 2010
lol
you cant make that knd of conclusons regarding how many hands are played each level.
they are not correlated – or can you provide proof?