A statistical exercise: short-handed weakness

[etotheipi]

Recently I have been talking to one of my friends about SnGs and statistics, and I decided to finally run the stats on my meek SnG track record. Mostly $20 SnGs on UB (a couple $10s and a couple $30s here and there).

For 76 SnGs so far my ITM is 46.8% and ROI is 40.5%. As I understand it though, n=76 sample size is nothing so I'm not on cloud nine yet. However, there was an interesting observation. I calculated for ITM=0.468, with a uniform distribution of the first, second and third places, my ROI *should* be 41.5%. I must not be getting my fair share of first and/or second finishes.

This inspired me calculate my win rates for these places. This was the result:
1st place: 14.3%
2nd place: 18.2%
3rd place: 14.3%

Before I did this calculation, my friend was saying how you should have more thirds and firsts than seconds, and I seem to precisely reverse that trend. I definitely missed the thread about this. Does this indicate a leak in my SnG strategy? I've always known that shorthanded play was not really my forte, and I'm not sure if I should be concerned.

Any thoughts on this? And if it matters what I can do about it?

Eto...

[john_]

I don't think the sample size is large enough. Maybe you're not taking enough chances when its three handed...its hard to say. Your ROI is really good though even though it is a limited sample size I don't know if I'd try to change what you're doing.

[Newt_Buggs]

look at the payout structure for SnGs, the biggest gaps are from 4-3 and 1-2 (a 2 buy-in gain). Improving from 3-2 only gives a 1 buy in gain. This means that you should be playing for 1st, not trying to sneak into 2nd.

Oh yeah, and prepare for the flames directed at the rest of your post /images/graemlins/grin.gif

[etotheipi]

Another side comment: (I tried to edit the post, but didn't work)

I wanted to point out that unlike everyone else and their mothers, I have no luck multi-tabling... this is why I only have 76 in my sample. Sometimes even playing two at once is a stretch. Maybe I just freak out unnecessarily when I have to act at both tables. Regardless, it'll be like 3 years before I get a thousand under my belt, so I have to work with what I have. Anyone else have this problem?

Eto...

[Freudian]

In the short term coming 1st and 2nd is a coinflip. I seem to get more 1st places when I put my chips in the worst of it than when I put the other guy all-in as a favourite.
In the long run I expect to get closer to normal results of course.

[viennagreen]

i don't understand why your friend thinks that you should have more 1sts and 3rds than 2nds.

you should have more 1sts than 2nds than 3rds... preferably, all 1sts. but different people have different goals i guess.... some people even like having more 3rds than 2nds than 1sts, but it's not for me.

[john_]

Obviously anyone would rather have more 1sts than anything else but given you could either have:

20 1st's; 20 2nd's; 20 3rd's OR
30 1st's; 10 2nd's; 30 3rd's

I'm guessing most people will take the second one. Basically tend to play for 1st when you're ITM.

[hummusx]

[ QUOTE ]
i don't understand why your friend thinks that you should have more 1sts and 3rds than 2nds.

you should have more 1sts than 2nds than 3rds... preferably, all 1sts. but different people have different goals i guess.... some people even like having more 3rds than 2nds than 1sts, but it's not for me.

[/ QUOTE ]

The idea is that if you are taking the risks you need to take in order to get first, you're going to end up busting out in 3rd more often than you would if you were playing more conservatively. The flip side is that if you are playing more conservatively and busting out in third less, you are also going to be shorter stacked on average when it gets down to two and end up in second more than in first.