Critique My Method

sdiinc,

Hats off for your effort and success, very nice job.  I think you may be onto something and I am not here to bash, just to understand and learn. My question is, how does this differ from a martingale system (which as we all know is a horrible strategy.) It definitely seems to be working for you, so I am just curious as to how it works differently...

Thanks and best of luck to you

PIT over CIN @ 200

Patrick, this thread is titled 'Critique my Method', so that's what I'm looking for. 

If you'll read my money management description above, I think you'll see that the strategy is different from the martingale.  Take a look at my post on 7/15. 

It is true that if we encounter a long losing streak the number of units we have at risk is greater, but it is not the doubling per event that is part of the martingale (as I understand).

I also talk about binomial distribution in a post above, so there is a potential risk out there that we're not accounting for, which I would happily accept ideas for better management.  Thus far our longest losing streak of 6 and the fact that we always begin with only 1 unit and recommend a 50 unit bankroll has resulted in a manageable level of risk.

sdiinc,

Thanks for your posts - much appreciated...

Have you implemented any sort of stop loss if you were to hit a long losing streak, or do you plan on betting however many times it takes to get a win?

Thanks again!

 

Well Zane, welcome to the thread.  I trust you'll look back on those early season losses and find some nugget of wisdom.  I know the sick, pit of your stomach feeling that comes with losing, and the important thing is finding something to call a 'lesson learned'.

I do have some experience 'pulling the trigger' and managing risk but I'm not perfect, so please feel free to offer ideas/suggestions or ask questions.

Betican, great question!  The short answer is no, but of course you have touched on a very real risk and something that has to be addressed.  When I started posting my longest loss streak was 3 games, we have now seen 6, and although I really like my own risk management strategy , we need to find a better solution to that.

I have been testing some other theories, but nothing definitive yet. I welcome your input.

i really like how your approach only wins back everything to break even...so all your units won are essentially the games that you hit on game 1. I think its a sound way to manage your risk.

For all the chases i have ever done, for whatever reason they all seem to stop after 6 straight losses. If you go 0-6, then u take it as a loss.
I know it is nearly impossible for you to do this, as there really is no definitive number of games for you...but some of the chases i have been playing, I already know all the plays before the season starts. One thing i do is i chase up to 3 games, and if i lose, then i take that amount that i lost and 'spread' it among the rest of the games left to chase. So if i just lost a series for a total of 240.00, and i know i have 30 series left till season ends...i spread the lost  (240/3 = 8).  If my TO WIN amount for 1 unit is $20.00, I would add that on and now I am betting to win $28.00. 
I know some people actually hate this method as the unit sizes may increase significantly during cold streaks and by season's end you could be betting to win $100.00 maybe. What do you think sdiinc?

DET over LAA @ 147

Thanks for the feedback.  I can't imagine how you can know at the beginning of the season what your plays will be, but perhaps you have a methodology that works, and that's all that matters.

My focus is on developing systems that could be implemented by someone who doesn't even know what the sport is.  My experience from the stock business has shown me that human intuition (influenced by emotion), is usually exactly opposite of what should be done.

If I understand your strategy, you are limiting your streaks to three losses, and increasing your subsequent unit size by enough to win back the losses over the season. 

My initial reaction is to ask you how you chose 3 games as your stop-loss as opposed to say, 4, which based on 50% is less than half as likely to occur as a 3 game streak (5% vs 11%).  From my own experience this season a 4 game losing streak is not uncommon.  This would mean that if you only changed your methodology to 4 games, you would reduce your current losses by a substantial percentage. 

One of the integral parts to my strategy is that the high ML keeps the risk low on what I think everyone calls as 'chase'.  (I try to avoid these terms myself because I find them to be imprecise and to evoke alot of emotions based on past experiences that may or may not be relevant to the strategy I use.)

Anyway, after a 4 game losing streak with my strategy it is more likely than not that your total investment in the streak would have been only 4 units and the 'new money' at risk in game 5 would only be 2 units.  With a 50 unit bankroll I don't consider this a substantial risk.

Thanks for replying and the valuable advice on the 3 and 4 game streaks. you bring up excellent points,  like why it is at 3 games...i suppose it is mainly because most series are 3 games series. The reason i know beforehand what to chase is because the system i am using looks into homestands and roadtrips of extended length.
What is the record you have and the net units u have won to date? And what is the most unit you have layed on a wager so far?
GL tomorrow!

TEX over BOS @ 125

My question was as much rhetorical as anything; the question is really, 'why?'  Why 3 games?  It presupposes an analysis that suggests this gives the highest probability of winning, as opposed to 2 games, or 4 games, or any other number.  If the answer is that this happens to be the typical series, I fail to see how it is that 3 is the 'right' number.  Again, perhaps you have some larger notion of a combination of factors that produces gains even though the individually measurable components do not.

YTD I am 70-68 on dogs with a 50 unit gain.  The largest single unit bet was July 28 and that was 6 units. 

I measure performance as much by efficiency as raw gain; we have a 50 unit gain on 138 events.  Because I recommend a 50 unit bankroll, this means I have a turnover rate of 2.76 (I have used my total bankroll 2.76 times to generate a 100% bankroll gain). 

I reference efficiency because we could make an average of 2 plays a day instead of 1, and that would allow us to make alot more money (a turnover rate of 5.4 and a 100 unit gain). 

Also, I measure efficiency because alot of other guys talk about a 50 unit gain but they've had to make twice as many plays to generate the same result.  In the investment business it's referred to as 'turnover' and it just means you're putting alot more at risk to gain a certain profit.

Okay good students, I promised I would begin posting NFL picks and I've started a thread titled 'Critique my Method (NFL) here for that purpose.  Enjoy!

FLA over STL @ 140

Awwwwwww....I love the smell of competence in the morning.

Thanks so much sdiinc.  Thanks for the patient explanations, and most of all for sharing your wisdom.

BAL over CLE @ 120

TEX over SEA @ 145

you said that aug was your worst month and so far in sep you are 2-6

my ?? is.. do you think that your system of picking underdogs is less effective late in the season due to other teams urgency to win??

Good pick sdiinc. Let´s get another one tonight and get that rally going!

3-6 now but i'm pretty sure he's still not in the red this month.

hey sdiinc, do you mind sharing your method for nhl last season as well?  or formula you used ??

Yes, August was the worst month with a 9 unit gain.  We had 10 unit gains in April, May and June, and 11 unit gain in July.  So, statistically it is a marginal difference.  This month we are up 1.5 units after 9 plays, so if it continues and we have a 4.5 unit gain for the month, it would be our worst month by half.

I think it is a mistake to look at 9 days and come to a conclusion about anything.  Even over the course of a an entire month you're looking at 30 events max, hardly enough to claim statistics as evidence for any conclusion.

Now, your final question does have some validity, but I'm not sure how to measure it.  I've had alot of suggestions in this forum and by email to consider other factors, but my system is based on statistics and not situational factors, which I'm not sure how to measure.  As I've said before, it 'makes sense' that some teams would try harder than others, and it 'makes sense' that this would result in favorites winning more often, but without statistical evidence that this change in behavior starts on a certain date, and is the case against underdogs in certain situations, and is measurable, I'm not sure how to implement it.

I will reassure you though with anecdotal evidence and say that last year my performance in September was not significantly different than the previous months.