Following on from How
Accurate Are Your Statistics, we shall now analyse the number of hands we
require in order to confidently assert that a player is too loose.
| VPIP | n |
|---|---|
| 40 | 369 |
| 41 | 258 |
| 42 | 191 |
| 43 | 147 |
| 44 | 117 |
| 45 | 95 |
| 46 | 79 |
| 47 | 67 |
| 48 | 57 |
| 49 | 49 |
| 50 | 43 |
| 51 | 38 |
| 52 | 33 |
| 53 | 30 |
| 54 | 26 |
| 55 | 24 |
The most common statistics used to measure a player's tightness are Saw Flop % (SF) and Voluntarily Put In Pot % (VPIP).
Poker Edge states that 33% is too loose, 18% too tight (source). This is for 9 or 10 player tables.
So, one way to find loose players is to find players who play more than 33% of hands.
From How Accurarate Are Your Statistics, we already know that:
p ± 1.96 x
Therefore, the lower confidence interval is:
L = p - 1.96
So
L-p= - 1.96 .
And thus
n=pq .
So, using the above, we can calculate the number of hands we need if we want the lower bound to be L, for a given p.
Table one shows some pre calculated results for the minimum number of hands required for an observed VPIP, given that we want a minimum VPIP of 35%.
For example, if Poker Office indicates that a player has a VPIP of 44%, then we'll need atleat 117 hands of data, if we wish to be 95% certain that the player really does play over 35% of hands.
From the Livetracker screenshot, we can see that Mr Elnur saw 49% of flops in the 45 hands present in the database. From Table 1, we can see that for players who've seen 49% of flops, we need 49 hands of data before we can be 95% confident that they see on average over 35% of hands. 45 is less than 49. So we cannot conclude that Mr Elnur is a loose player whom we want to play against.
However, these figures need to be taken with a pinch of salt, as there are some problems with the assumptions.