The Bankroll Building Project (Guide to Micro-Limit SNGs)
by Chris "Fox" Wallace

Some ROI Simulations
posted by fox on February 26, 2008

After a rough day yesterday where I only cashed in one out of five SNG’s I played, I decided that today was going to be a research day. I checked Sharkscope for my exact stats for the project to start out (there are some cash game winnings, rakeback, and a small MTT finish that you can see in previous results posts, but Sharkscope is perfectly accurate on my SNG results as far as I can tell).

With only 150 SNG’s, my stats are certainly not guaranteed to be accurate, and I would have a much better idea after 500 or 1,000 trials. The sample size is still within reason even at 150, so I decided to use the ROI simulator (no idea where I downloaded this thing a few months ago, probably twoplustwo) to see what my variance might be. I had to round off slightly to get the simulator to work, but it was close enough anyway. I also told it my results were form $24 SNG’s when half of them were from lower levels, but I had to pick a level for it. I don’t actually think I would maintain a 34% ROI if I just played at th $20 level, though I might be able to pull it off at the $10 level. Here’s the screenshot of what the simulator told me, and then I’ll explain a little bit about the stats and the variance.

A few notable pieces of information we get from simulating 1,000 games 1,000 times -

90% of the time my ROI over a 1,000 games will be between 26% and 42%. That is VERY comforting, given how high I know the variance can be in SNG’s.

5% of the time I will have a drop of more than 23 buy-ins over the course of 1,000 games. Not so comforting anymore. After a drop of 23 buy-ins, I imagine almost anyone will seriously start to doubt their abilities. I know I would be pretty stressed, even though I know I can beat the games and that my opponents are making significant mistakes.

95% of the time my lowest point will be down less than 10 buy-ins from where I started. That’s nice to know.

Given the ROI numbers, I should be up between 200 and 400 buy-ins at the end of the 1,000 games. At the $10 level, where these numbers might be achievable by a very good player, that means between $2,000 and $4,000 in profit from this chunk of play. 1,000 games, assuming 8 games per hour and 40 hours per week, would take a little over three weeks. Not bad for a game you can play with a $200 bankroll.

Of course there aren’t any players of my skill level that are grinding out thousands of games at the $10 level. A player who can beat the games this badly at the low limits will move up quickly and try their hand at the higher games, possibly moving up until they find a level that they can not beat, known in the business world as The Peter Principle.

I’ll be doing some research in the next two days about whether the cash games I can play with a $1k bankroll are better than $30 SNG’s and then I’ll be on to the next step, whatever it is. I’m on vacation from the 27th to the 6th, so no posts during that time, but if I don’t post before then, I’ll see you when I get back. I’ll have the computer in the shop while I’m gone as well, hopefully returning to a slick quad-core system or at least something that can more easily handle all the software I’m running and the two big monitors.

A Nice Bounce
posted by fox on February 25, 2008

Played six $22+2’s tonight, with finishes of 5,3,2,1,1,1 for a nice bump in my ROI. The new spreadsheet I started a few days ago has a 23% positive ROI now, and my overall ROI for the project is back up to well over 30% now too. Current bankroll is $1,027.80, and I have some rakeback coming as well.

I would have played a lot more than six tourneys tonight, but I was having some issues with my computer where it was flashing the ables and not always showing the check boxes and action buttons. I have seen this problem occasionally before, and it happens on Stars as well. I know I’m running a lot of software, but a nice dual core 64-bit machine running XP really shouldn’t have this problem with just four tables and some tracking software and AHK scripts. I need to find out what the problem is and maybe buy a better video card so that I can really sink my teeth into this project and get some games played.

I’m basically ready to move up to the $30 level now, though I need to review cash game win rates again and make sure that there isn’t a better place for me to be than $30 SNG’s.

Some Stats (and a rough day)
posted by fox on February 23, 2008

After a rough day yesterday where my only cash was one third-place finish in six attempts, I kept the rough streak going today with six straight losses. I almost recovered all of my losses for the day with first and second place finishes in my final two SNG’s of the day, leaving only slightly down for the day.

Overall ROI for the project 31%

Current Bankroll $776

I’m looking forward to PokerTracker 3 launching with PAHUD so that I can start using for SNG’s. I think it will use up less processing power, which will be big for me, because playing more than four tables with everything I have running to make the playing easier is starting to give my computer some problems. Does anyone know why my page file usage is in the yellow even wen I don’t have hardly anything running at all?

Back to the Grind
posted by fox on February 22, 2008

A quick update and then some thoughts -

I’m back, feeling healthy and ready to play some cards. I bought a new keyboard a few days ago because the mutt spilled water on the old one, so please forgive any typos that I don’t catch while I get used to it. I also bought a 24″ 1920 x 1200 resolution monitor and I am absolutely in love with it. How did I not buy one of these things before. Coupled with the 1600×1200 next to it, and merged as one desktop, I can play ten tables without any problems. At lest not any space problems, but more on that in a moment.

The monitor was on sale for $279 at OfficeMax, and it looks so good that I may go back and buy another one. Awesome deal, and so much better than the monitor it replaced. I highly recommend all serious players drop by your local OfficeMAx and get one of these things if they still have them in stock for $279, it will make your poker life so much easier.

Of course you have to be careful how you use all that space, it’s easy to play too many tables, and I’ve preached against that often in columns and blogs in the past. I played six $24 SNG’s a few minutes ago on the new monitor, just to see what it was like. I took some beats, and only managed one third place out of the bunch. At last a small portion of the bad finishes was due to playing to many tables and being unable to concentrate on my opponents with too many tables beeping at me.

As any of my SNG students will tell you, my game relies heavily on finding the right players to abuse and robbing them blind during the mid-levels. That is much tougher to do with six tables than with three or four, and I ran into some players who called with a much wider range than I expected. I was called by hands like AT and KQs a few times when it was clearly a bad play on my opponent’s part, but the difference between a serious pro and a weekend warrior is that I don’t complain about their play, I look at my own mistakes and try to fix them instead.

It doesn’t matter that my opponent was making a big mistake calling my push on the bubble with KQ, it matters that I pushed into a guy who would make that call and I didn’t know it. If I had been playing four tables instead of six, I may have noticed that tendency in my opponent and avoided the risky push. Don’t let yourself fall into the habit of complaining about your opponent’s play at times like this, there is nothing you can do about their play except make them better. Just look at what you can do to fix it and find ways to take advantage of their mistakes.

It’s also just my luck that I started a new SNG tracking spreadsheet for the project today, so my ROI is currently -72.5% hehe. Ouch. My ROI for the project overall is still over 35% and I’m sure I’ll do just fine, a small blip like losing 5 out of 6 is not a big deal at all, and in fact it’s something you could expect to see once a day if you played SNG’s for a living. I may even see a drop of over 20 buy-ins at some point in the project if I keep at it for awhile.

I will be going back over the SNG’s I just played to determine if I really made bad plays or just got unlucky, because I want to be sure about things before I give up on the idea of 6-tabling them. Playing six tables constantly would earn me about $6 an hour in rakeback, and at a 10% ROI I would be making around $25 an hour playing as well. The question is whether it is significantly more profitable than playing 4 tables given the likely drop in ROI from the extra tables. It would have to quite a bit more profitable, because it’s definitely more stressful and more hassle, even with the big new monitor.

My other option might be a hybrid scheme, possibly playing sets that start with six tables so that I am usually down to a smaller number once I get shorthanded at the tables that are left. I’ll have to check into that, but previous number crunching made that seem like a bad idea in terms of hourly rate.

I also thought about playing 12 tables at a time and just not worrying about getting any reads other than what PokerTracker and PAHUD give me, but the variance would be higher and my brain would really fry with that many tables running. I don’t know how people manage it. I do know there are people playing upwards of 20 tables on Stars, but I also haven’t found anyone who is a serious winning player while playing that many tables, so it’s really just something you do to prove you can fit that many tables on your screen or impress your pals on a forum.

Almost three weeks since I played a sit and go, but these were fun, even with the bad results. I’ll be playing some more later tonight or tomorrow.

Staying Motivated
posted by fox on January 20, 2008
I’ve helped a number of students with the issue of staying motivated even at stakes where the money doesn’t feel like it’s worth it, and I know it can be tough. I face that myself sometimes with this project because I have so little time to play poker these days and it’s tough to play micro-limits when I do finally get a chance to play. Playing at a little higher level over the past couple days for a few hours hasn’t made that any easier, though it has been nice to make a few bucks.

I played some $10-$20 No-limit on Bodog last night, along with some $30-$60 fixed-limit, and was up over $1,000 on both tables when I left. When I combine that with the higher level SNG’s I played a few days ago, and I’ve made around $4,500 in a few hours of play. Makes it tough to go back to $20 SNG’s, but it’s also nice to know that I won’t lose it all back in a hand or two and be grumpy for the rest of the day, so maybe the lower stakes will be nice.

I really do enjoy the project, and within a week or two everyone will understand why I’ve been too busy to spend much time on it for the last month or so. Really fun news coming very soon!

Fixing those sharkscope stats
posted by fox on January 17, 2008
Because a very small sample had hurt my sharkscope stats, and more and more people are reading this blog, I figure I ought to actually fix them. I played three sit and go tournaments yesterday, one $330+20 and two $110+9’s. I won the $330+20 and took a second in one of the others, for a nice boost in the sharkscope stats. It will take a little time to build up from the short rough patch I had at the $500 level (damn you SamEnole) but I’ll get those stats looking better overall so there aren’t any questions. The $1,600 I made yesterday put a nice bounce in the sharkscope graph, but those obviously don’t count for the project. Playing with half my bankroll on the table for one buy-in would be a little more risk than I usually take.

My Sharkscope Stats
posted by fox on January 14, 2008
In a post on pocketfives earlier today a poster questioned my Sharkscope stats on the account I have been playing on for the Bankroll Builder Project. I thought I should just clarify a bit, lest anyone else attack my credibility or my skills without a good understanding of what they are looking at.

First of all, that account is vulpesvulpes on Full Tilt.

My average ROI is quite good, though my total income is a negative number. How did this happen? I played a few $500 sit and go’s, and had a short rough streak. After that I started the Bankroll Builder project, and began playing $5 tourneys instead. A small rough patch of less than 10 buy-ins (in less than 50 games) makes for a very ugly graph, but says nothing about my skill level. A 10 buy-in swing is very common, and a 50 buy-in downswing is easily possible for even a solid player, so I wasn’t worried about it at the time. Of course I didn’t expect that someone would check my sharkscope stats and assume I’m a fish from a very small and common downswing either.

So there you have it. Any research on variance in SNG’s, especially if you use the roi simulator from 2+2, will help you understand variance in these tournaments. Notice that my average ROI is still excellent, but a few hundred sit and go’s is really not a large enough sample to be sure. Either way, I’m running at a 40% roi since I started the project, which is the slow climb for the last 125 games or so.

Running Awfully Good
posted by fox on January 10, 2008
My hot streak in the sit and go’s continues! I’m running at a 42% ROI, and it feels like I’m taking a lot of beats, but I really don’t think an ROI that high is sustainable. Especially when my ROI is almost as high at the $20 level as it was at the $5 level. After cashing in three out of four tournaments a few minutes ago, the bankroll is now at $920, and I think I may have missed some rakeback cash as well, so it could be a little higher than that. I am eager to move up to the $33’s and see how the play is there.

The sample size for that 42% ROI is 123 tournaments, which is certainly not big enough to say that it can be sustained, but it’s big enough to tell me that I’m playing pretty well on top of the nice run of cards I must be seeing. I’m off to play some micro-limit no-limit games because some members have been asking for videos and articles on them. Might have time to play a little more this week, hopefully get into the $33’s and get a few games logged and get the finishing touches on the $20 SNG guide.

VP$IP According to Blind Level
posted by fox on January 7, 2008
While I haven’t played enough for most of my pokertracker stats to be useful, and I also don’t import my tournament summaries, I think this little graph might be interesting to many SNG players. A few hundred more tourneys would give a little more accurate numbers, but these seem to be fairly indicative of the style I teach and recommend. I also haven’t seen this sort of breakdown posted before, though I’m sure someone has done it somewhere.

Looks like a fairly steady climb as the blinds get higher, until the sample size gets too small at 400/800. I do play a few more hands than I recommend in the guides I’ve published here on the blog, but that’s because I’m a very experienced player and I have a huge advantage over the typical micro-limit player after the flop. For most players I recommend about 3% tighter than I play except during the first level. During that first level you may even want to be a little looser than I was, especially in the $5’s, because you have a chance to break a weak player in the first few hands.

That may seem strange, but for the typical player those chips they can earn in the first level are worth taking a risk of going bust early. As your skill level increases, and you have a larger and larger advantage over your opponents, and you want to gamble less. In the lower limits I believe that my advantage is large enough that I can afford to give up small edges early in order to make sure I get to use my edge later in the tournament when it is larger.

SNG Madness on Full Tilt
posted by fox on January 5, 2008
I’m hoping to get some SNG’s played this weekend as part of the bankroll builder project, but I’m not sure how much time I’ll have. The SNG Madness Promo on Full Tilt looks pretty good for SNG players, though you have to be at least 10 tabling to make the leaderboard. The raffles could be useful though, and it certainly has a bunch of extra people playing SNG’s. After some good luck in the 9 $22+2’s I played today the bankroll stands at $711. I plan to move up to the $30’s at $1,000 or so. The $20 SNG guide will be done this week some time.

Sit and Go Madness!!!
posted by fox on January 4, 2008
Full Tilt is once again running their Sit and Go Madness promotion, and it’s pretty easy to win the leader boards if you just play 4 to 6 tables of $22 SNG’s. You have to win a two hour chunk, but I haven’t figured out yet if it counts SNG’s that you started during that time or SNG’s that you finished during the two hours. For each two hour block you can make leader board money, and they are giving away raffle prizes all weekend too, so it’s a great time to be playing on Full Tilt.

Unfortunately I won’t be able to play all that much, but I should be able to get some sessions in and get some stats for the $20 level. I’d like to get the $20 guide written this weekend as well, we’ll see how that goes.

If you are playing on Full Tilt you should definitely be getting rakeback, and if you don’t have an account yet you can get a max percentage account here - FULL TILT RAKEBACK -

NEXT PAGE >