Kelly/staking-question

[Cashbabe]

Hi, I am pretty new to sportsbetting (compared to most of you guys in the ThinkTank at least), meaning I just started a little five years ago because of a friend/colleguae of mine who have been handicapping successfully for some years, and introduced me to it.

Did a lot of bad mistakes the first years, and lost a fair amount of money, but the last year my results have really picked up after I started reading through the whole ThinkTank, and ordering most of the recommended books. I read everything I canI now think of the losses in the beginning as necessary costs for the knowledge I now have. I have not been repeating the mistakes again, and I keep detailed records of all my wagers and results to filter and improve on what strategies that seem to work out best for me.

However I have been starting to think of other strategies as for my stakings. Until now I have only been flat-betting (3%-5% of my starting bankroll). I have been using a very selective strategy with fewer games (about 1 play per day), to keep variance lower. I know I might be giving up on some potentional profit with this strategy, but it has felt like the most secure way during my newbie learning period.

Since the past 12 months have showed very stable positive results I think I am ready to use a more optimal staking strategy.

I have a program that simulates diffrent types of staking strategies with the exact records I have of my previous picks. I see that the flat-staking have cost me a lot of profits, even though it feels pretty safe to use. I currently have a bankroll of about $20,000, and I want to grow it slow and safe rather than risking losing it all.

Since I am focusing only on a few plays, the simulations showed that if I had flat-bet 4% of current bankroll instead of flat-betting s set percentage of the initial bankroll, I would naturally have earned a lot more in 2010, and without any bad downswings either. I have been thinking of betting 3% of current bankroll, but never lower my bets, only increase while I'm building the bankroll to 50K. Is this too risky? I see most of you say you should recalculate every day, both increase and decrease, but would you think it's okey to just increase by each peak during a temporary period of building the bankroll?

Now my next question is regarding Kelly. I have read most of the threads about it, and I find it very interesting. However my problem is I don't feel secure about the edge for each pick, I just know my overal edge. So if I understand most of your threads correctly, if I only know my overall edge, I should not be touching Kelly at all? If I were using 1/4-Kelly, it would be about the same results that I got if i simulated 4% flat of current bankroll per bet.

Long thread, but hopefully someone can give me an advice as to how I should proceed with my staking-strategy in 2011.
I'm leaning towards 4% of current bankroll, never decrease until I reach $50K BR, and then keep my few, top plays to 2 or 3% max. I can accept a little risk in the beginning to build the BR, but I am thinking long term goals in years, so I'm not in a hurry, I just want to do it right. I have a full income from my company on the side, so I will not be touching the BR for living expenses.

Thank you in advance for any help and advice, and not to forget: Thanks a lot for the incredible resource of knowledge inside the ThinkTank!

PS: Sorry if there are spelling errors here, english is not my first language

[Metanoia]

Hi Cashbabe

I can give 2 advices based on my experience, i think you should recalculate every bet because that way you take better advantage of a winning period, and you protect yourself more in a losing period

About Kelly, i believe it's probably the best way to increase a bankroll, but also the fastest way to destroy it if you don't know what you're doing, you have to be very sure of your edge and believe me we all have been guilty of thinking our edge was better that what really is

Just my 2 cents here, but there are a lot of other people in this forum with a lot more knowledge for sure

In what sports to you bet most?

[Cashbabe]

Hi Metanoia,

thanks for your reply! Yes, that is the reason I am doubtful I should even think of Kelly, because as mentioned I don't know the edge of each pick, just that my overall numbers are good. When I read "Fixed odds Sports Betting" by Joseph Buchdahl (good book btw), he assumed that the edge for each Kelly-bet was exactly the same, just to make things easier when doing the Monte Carlo simulation.

But is this something one can do in real life Kelly-betting? To just use the overall edge for all picks, or sort the picks into groups for each odds, for example my edge for 1.91-odd picks in one group and my edge for 3.0-3.5-odd picks in another group, as the edge is very different on those groups, but both show good results.

But then again, maybe it will be much less hassle to just use a set percentage of the current bankroll and recalculate.

If I were to do a light version of Kelly, I would assume a lower edge than my records have shown, and probably use 1/4-Kelly. But using the same edge for all picks with different odds would not be a good idea, right? That is mainly what I was wondering. I think it would give some scary downswings when the large odd picks misses...

I don't need a heart attack

[Metanoia]

I would keep things simple, recalculating bet by bet.

I bet mainly on soccer and it works very well

[Justin7]

3-5% is very aggressive. Going up from there sounds like a recipe for disaster.

I'd suggest you limit your max bet to 1% on any side/total until you have had 2 winning years in a row.

[Cashbabe]

Hi Justin, thank you for your response, I have learned a lot from your posts in the tank I will keep your advice in mind!