Do my third and fourth games hurt me?

[Skipbidder]

Recently, I've been playing mostly limit SNGs at 50+5, two or three at a time. (I play 100+9s when they are available, which is not that frequently--a problem with NL vs. limit, I think)

I used to play 20+2s four at a time, and felt comfortable doing it, but there are often not that many limit 50s or 100s available. I'm wondering about whether adding a low game will have a significant enough detriment effect to my higher games that I might actually lose EV.

Is there a good way to figure out whether or not adding a 20 (or two) to my 2 or 3 bigger games is worth it? I don't really know how to best gauge the expectation effects of adding a third or fourth game. (I've got pokertracker, but I can't figure out a good way to use it for this purpose.)

My ROI numbers in the 30s really, really stink compared to everything else from 10s to 100s. (I don't know if this is a chance fluctuation or not, but it seems to be steady over time.) If those numbers are right, then I'm significantly better off playing a 20 than a 30.

Thoughts about whether or not adding a 20 is likely to be worth it?

Not sure if it is at all relevent, but if you need them, here are ROIs since January 1:
100s (n=26), ROI 27.03%
50s (n=124), ROI 40.76%
30s (n=72), ROI 17.42% (This is actually a significant improvement over last year's 30s, which ran in the 8-9% range. I have no idea why the 30s are so much lower than tourneys above and below.)
20s (n=101), ROI 59.32% (These were mostly 3-4 tabling, but didn't include any bad runs, thus I consider the number seriously inflated)
10s (n=48), ROI 36.36%

Thanks,
Skipbidder

[raptor517]

well, i probably wont be the first one to say it, but your sample size is entirely too small. even after 400 tourneys, after a bad stretch, my roi jumped 10% in just 20 tourneys. so i wouldnt base your strategy TOO much based on your numbers. just play whatever you are comfortable with.

[Skipbidder]

I worried about including the stats at all, but in the end thought that some information would be better than none. (The decision to use numbers since January 1st was entirely arbitrary, based mostly on the fact that I had a printout in my pocket.) As much as anything, it was to be an explanation of why I'd add a 20 instead of a 30.

I'm unlikely to play tournaments at a rate any faster than this, thus I didn't really think I'd have any way to empirically compare expectation for two 50s vs. two 50s plus one 20.

Perhaps a gut feeling would be the best way to handle this situation, but I don't trust my own gut feelings on the subject.

[YourFoxyGrandma]

You were the first one to say it.