No matter what, don't post the screennames of anyone, ever! Simple as that.

_______________________

That said, let's get to the content and discussion.

This Thread will (or should) be about Game Selection as a whole in SNGs. This includes Methods, Guidelines, Rules of Thumb, Math, Theories, (legal!) Software, Websites, etc. etc.



I'll make the start with how I gameselect:
I'm playing the $12-$36 turbo 9-mans on FTP and use a simple colorcoding system to decide which games I want to join or pass up.

- big fish. -5% ROI or worse over a decent sample
- fish. -1 to -4% ROI over a decent sample, has a bit of a clue.
- breakeven. can range from -2% to +2% depending on the swings and their graph
- slight winner. +1% to +3% ROI depending on swings and samplesize
- winner. +3% to +5% depending on swings, sample and avg. stake
- good winner. +5% to 7% over a good sample (5k min.) and decent avg. stake
- solid reg. 6%+ over a big sample and good avg. stake

All of these tags are based on sharkscope results if I have no additional info. If I see someone, who already has a marker, do something good or bad they can move up or down 1 or more classes. If I see an unknown do something really dumb, they'll get a red or pink marker depending on how dumb it was. To date my "database" contains around 6.000 players.

I play in sets or continously depending on the quality of the games and up to 12 tables. I gave up just registering for 12 tables and take who ever joins, I am just not good enough to do that. So I register for games that I feel I'm +EV in and those usually have at least 5 players registered already.

I usually pass up games with:
- 1 solid reg and at least one breakeven player
- 2 good winners
- 1 good winner, 1 winner and a breakeven player
- 1 good winner, 2 slight winners
- 2 winners and a slight winner
- 2 winners and 2 breakeven players
- 1 winner and 2 slight winners
- 3 slight winners
- 4+ breakeven players

I said "usually" because there are exceptions of course:

Fish
Fish are able to make these games playable and profitable, but it's too dynamic to make a good rule of thumb out of it. Usually a big fish or 2 normal fish compensate for a good winner, a normal fish compensates for a winner and so on.

Keeping games going
If I have loaded a full set and bust out of one tournament i'll try to fill that gap or if bust or win some tournaments and I feel like i'm playing well and I still have some games in the earlier stages (like 30/60) I will load additional games, but with lower than my initial standards.
This is because good gameselection needs time and so do the games that are still running, but the latter are more important. A bad gameselection decision costs me less than a bad ingame decision. So i'll load some games that are probably less +EV than the games I usually join. Plus they'll also help to keep me focussed.

Last but not least, the link to jhub's blog entry about Game Selection:
http://jhub3000.livejournal.com/84368.html



Thats all I have to say I think.

I hope many follow suit and we'll get a decent discussion going on how to gameselect optimally. And in case you forgot: NO NAMES!

nice post. generally good guidelines. i just wish they were applicable to stars $16+, but you're high if you think you can consistently find games that don't have at least a couple "solid regs" or "good winners."

very nice post..

not that true, when i played them i would normally only have 1 or 2 breakeven regs at my table at most

simply load a brand new set, if a really good reg loads on top of me then unreg and start agian. they fill up so quickly there's no excuse. unless they are 2+2 then you play them and spite call ftw

its actually the other way round if you play high you will be luckily to find one fish at the table

Great post.

I really have a hard time understanding why a lot of ppl choose to play a game filled with regs rather than wait 2-3 minutes and load some games below the reg infested ones.

It's mind boggling at times. Ppl really need to forget about their ego's and think about long term profitability.

agreed that table selection is very important, and regs in general do not table select enough. The colour coded marking system in the lobby is a great idea from full tilt.

youre site and stakes selection seem pretty good if you can pass up on this many games and get enough traffic to multitable. At higher stakes on my site the only choice is to play in a table full of regs, play a different buyin, or not play at all. Also why is there no colour for players who run below -15% over a big sample as they actually will make up for a good reg being at your table.

Agree. Nice post Jurass, but I really think you're setting the bar too high with your expectations.

Online poker in all forms is rife with regs at all levels. Just go and try $50NL and $25NL full ring, it's flooded with them. Online poker is getting to the point now where there isn't a whole lot of soft money around, and the way to win now is tp focus your efforts on improving your game and being better than the next reg, which imo isn't that hard since there's tons of really meh regs around at the low stakes games. I haven't table selected once yet on FTP and have only seen about 3 or 4 players who I am remotely concerned about. Sure there's a few b/e poxy rakeback grinders, but who cares about them, they are so weak passive there's money to be had from them.

FWIW Stars 16s and above are far far worse regfests compared to FTP. And at the 60s levels, it's just one massive pool of regs of varying ability swapping $ back and forth.

1. I def use more fish colors than reg colors. I look at fish roi's and separate the ones with -20%+, -10%+, or something like that and for regs I just code marg- regs, breakeven/mar+ regs, decent regs, regs to avoid.

since u have a SS account i would def filter by type of game ur playing (i'm guessing full ring) and look at how they are performing stake by stake. I know my SS roi is like 1% b/c i donked around at superturbos/micro stakes a lot when it's actually 10%+ for the games i play/stakes i play.

Following Jurass's guidelines it would be almost impossible to get any volume at the 16's but if ppl just started off by avoiding loading into a game already containing 3 decent regs , that would be a start.

If the lobby has multiple games already with 3 regs in them , and if I quickly check them and see none are real losers , then I just load under. Not really difficult.

I really don't get how being the 4th reg in is a better alternative to loading a few games below and waiting a few more minutes to play.

Nice post. I think one leak a lot of game-selecting regulars have is not worrying about registering with breakeven players, but avoiding good players like the plague.

Remember, a BE player covers both his entry fee and the rake. So, in a $24 SnG that means he takes at least 12% out of the prize pool (entry fee=11%, rake =1%) whereas a good player with a 5-6% ROI takes 17 or 18%. So, 2BE players is much much worse for you than one good player.

Also, the different between a "kinda good" player and a "really good" player is pretty minimal. A really good player at the $24s on FTP has like a 6% ROI and takes 18% out of the prize pool. A kinda good player takes a full 12-13%.

This is why I don't really color code that much. I have one color for regs from BE to good and another for extreme fish. I then try and play in games with no more than two guys coded as regs.

Good Post on Game Selection

Another great post on how much regs hurt your ROI by CJ Saunders and why game selection is a must: http://forumserver.twoplustwo.com/36/stt-strategy/lc-500th-post-game-selection-how-much-regs-hurt-your-roi-411194


My game selection process

- Unfortunately I'm red-green colour blind so cant use any colour coding system like Jurasstoil. But both Stars and FTP (not 100% sure about FTP) let you see any notes you have on players in the tournament lobby. So I use player notes
- I am much more stringent and nitty than Jurasstoil (see post linked above to see why) and basically dont like playing any games with someone who has any clue what they're doing (even if they're break even). Since I play lower buy ins I dont see the point in playing with regs as there are plenty of fish out there.
- I make notes on players - sometimes putting in sscope stats - but mainly a small note that the player knows what they are doing: 'teg' ('tight early game') and then I have 'dkp' ('doesnt know push-fold') or pw ('pushes wide'). If someone is teg and pw I basically dont want to play in that game with them since it's much better to have a teg who blinds down than pushes wide. So when I say 'reg' I basically mean anyone who has a reasonable clue about STT strategy.
- I register for my set (20 tables) and as a rule I NEVER EVER register on top of another game. That is I always open 20 new tables that way I can monitor and control who sits down and there is less chance of me registering on top of good players. The way I see it is if all good players had this rule no-one would ever be playing in each other's games. As I mentioned, I play fairly low level and the games run often enough that this wouldnt be a problem. There is one exception I have where I see a game totally full of fish but this is becoming rare nowadays.
- I then have 20 tables sitting at 1/9 to monitor. If they start going 2/9 2/9 2/9 etc I look to see who is registering on top. If I have no note on them I shark scope them. But generally if they are playing >6 tables I will un-register from all these 2/9 games and re-register at the bottom (open new tables).
- If the person registering on top of my games has registered in one that is already at like 4/9 I wont un-register from that one. The main reason is a game that is 2/9 is much more likely to have at least another 'reg' register on top. Where as a 4/9 usually fills up to 9/9 quickly.
- Once I have most of my set loaded up (>10 tables) I stop the above process as it is too time consuming at this point. However if my tables have all just gone 3/9 3/9 I will un-register these and re-register at the bottom. I am now even reconsidering this since it is so time consuming and tilting when you're in the middle of a set to do this. It also makes your set last much longer.


Now it might seem I definitely go into over kill to avoid 'regs' and game select. However I feel it is beneficial to my ROI even if it does affect my hourly (game selecting takes you more time to play a set).

Sometimes the process doesnt work and I still get a bunch of tables with average reg's. I have no problem spite calling guys who register on top of me and saying so in chat .
I also openly berate someone who has registered on top of me when we get HU ('Wouldnt you rather be playing a fish HU right now? I know I would').

That's all from me.

I only play 5.50 STT SNGs right now on FT. Often there is not a single good player at my table but its not unusual to have 3 decent break even players. Usually the ones who are solid winners are on ten tables and I find those players are no problem at all. You just get very straight, correct, honest poker from them. They play with their cards face up. I usually only try mark the donks and I mark them based on bad plays they show at my table. I mark them green as in GO baby GO then I'll put a very short note like calls any raise or plays flush draws . Good players who I notice I mark in red. One day I had 4 very good players all in a row to my left.

For me the hardest part of the game is ascribing what type of stupid play each individual fish is prone to this game. For example some donks never show down with second pair but they will pay any price for a flush draw.

Is this on Stars? Do Stars now allow use of Sharkscope? A couple of years ago I got a warning about visiting Sharkscope website while the Pokerstars client was open and how it was on the prohibited software list and against their TOS. Have they now changed their policy to allow using it to look up players?

nice post , thanks ! and the blog is awesome too !

following this kind of chart is a volume killer though . so don't know what's the best ...

Yeah does SS even work on Stars anymore?

It was only for a few days that Stars were enforcing that policy.

There really is no problem using SS on Stars.

QFT. The people who do this, some of whom post regularly in this forum, might as well be playing roulette for a living.

Maybe learn and incorporate different games? Wouldn't we be able to table select AND achieve volume if we added let's say 6m or multi table SNGs?

mixing different type of games at the same time ain't easy at all imo (i ve tested a bit)

Personally I'm glad they do. Which is why I'm not super nuts about this thread. But my guess is the guys with the ego to keep playing with all regs will keep doing it either way.

Bugger. I've missed out on all that info for years now
I must have just caught the wrong time window to use it.

Good post this is definitely not something I do enough of. How do you go about looking up players? Currently I don't have a sharkscope sub and I just look up players that are at my table that I have played a significant amount of hands with until I've used up 5 searches. I also tend to just register first and hope that other regs see me and decide to not play with me. I also play 6 max which fills very quickly which is not ideal.

I know, but i'm not good and lack confidence in my game (rightfully tho!), so I have to gameselect quite heavily to stay profitable. But I'm looking to refine my gameselection methods to make them more optimal.

I do. I have setup some filters to get the results i need. For the quick searches i use a simple 9man filter that is filtering for NLHE 9man turbo SNGs, if the sample is big enough i'll refine the search for a certain buyin range or even a specific buyin.
Yesterday I found a reg with 25k games under his belt, a solid ROI and a purple marker. I checked him again (the marker was from before I had a SS subscription) and found out that he struggled to breakeven in turbo sngs over more than 8k games and his big ROI and profit came from normal speed sngs.

Thanks first of all

I'm not saying "Use my guidelines!". This thread is intended to discuss the best way to gameselect. The guidelines I use meet my preferences and my self-assessment (See my reply to NJD's post above) and are far from perfect. My Games per Hour are decreasing every month atm and its starting to worry me. Thats why i'm looking to tune my gameselection methods and rules.

Note that there is a big difference between the games on stars and FTP. I'm not talking about the number of regs or fish, but about the Rake. You won't be able to achieve a 12% ROI at the 12s on FTP as opposed to the 16s on stars. So, the ROIs are considerably lower because of the higher Rake%. As a result the amount of winning players (players who beat the rake) is far higher on stars eventhough their skill would make them losing or breakeven players on FTP.

I have a Sharkscope Subscription (150 searches/day).

6max is even tougher to gameselect, because already 1 winning reg can hurt your ROI quite significantly. I've never played 6max and haven't really thought about them either, but faarcyde just released a pretty good video about gameselection for 6max sngs on stoxpoker. So if you have subscription go check it out.

___________

I selected 6 22s on FTP (3 yesterday, 3 today):




Which of these games would you join (replacing one of the winners)?

pretty hard to do on stars at the 16s. Game selection is important and something I practice. If one of the better regs has fired up tables I will load below them. If I fire up a load and they join then I don't unregister. When the games are good I add a few more tables than I normally play.

I think time of day you play is more important. For me its a choice of playing when all the sharks are feeding and the fish content is highest or playing when its quieter with a lower calibre of beter players and less fishier fish. I personally opt for the latter

This is probably a horrid way to go about it, but I usually try to join games that have 1 person in them(preferably a good reg), that way I start registering and it says 2/9 for like 10 games...my goal is to scare off as many regs as possible....I would rahter join the best reg at the 16s when hes first and im second than 3 mediocre regs....cause a lot of mediocre regs avoid just te top players...Which means overall these games could very well be softer.....