Critique My Method

with the lehman news... I wouldnt want to be holding any stock right now.  IMO

I'm more of a value guy.  I'm also contrarian so I'm sitting on a lot of cash right now, buying stuff on the dips. 

Energy has been good to me this year but mostly it was oil-related in the first half of the year.  I like pharma and Asian/LATAM telco related as well.  Buy stuff that is undervalued/oversold and works well in tough markets.

So people are emailing you to get info about the betting system and then we who read the thread gets info about stocks. Maybe it can be the other way around since i really like this thread?

I have some info to share about money management. Instead of just trying to get my losses back starting with the second bet, i bet to win a smaller amount on the second bet as well.
I had a win yesterday since i bet to get my lost unit back from the first bet and then added half a unit for a smaller profit. If it doesn´t work on the second bet, then i start to bet to just get my losses back starting with the third bet. Sure, it means that you need to bet a little more to get back to even if the first two bets fail but not much. It´s been working so far and i will continue doing it since i don´t see it make a big difference in risk.

SEA over KAN @ 135

Obviously it was a mistake to post my email.  I get complaints on email and people on the thread complain as well.  Excuse my first experience with managing a thread.  I'm beginning to think it's not worth the pay.  Oh, wait...

As for the progression, I think the approach you describe is too risky-you'll have alot more at risk on a 4 or 5 game losing streak than I will.  The reason I use the approach I do is it allows me to chase a loss with the least amount possible. 

A recent email exchange which may be informative (my responses in red):

First off i'd like to ask you about your baseball system. 
1)  You start with a 50 unit bankroll and you mentioned that you were able to double your bankroll every month.  Did you mean starting from 50 units, you doubled your bankroll to 100 units? Or a ROI double ? Where you wagered X amount and achieved 2X back?

I shared this before I came to the conclusion that I was betting too much per game as a %.  In reality after I've adjusted my unit size to 2% of my portfolio, you would be looking at something more like 20% gain per month.  Still great for me, but some guys have mocked that, and they're using unit sizes of 5%, which would translateinto 50% monthly.  After I ran some intensive tests I concluded 5% was just too much for me, and if I can get 20% a month, I can retire in six months and do sports investing full-time.

2)  You need both formulas to agree either positive or negative right? But if it is negative that means that it is predicting the road team to win right?  So if you are playing the home underdog....but it comes out to be a negative number, suggesting a road team victory, does that cancel the play?

I do look for agreement, but I frequently accept disagreement if it is slight.

Because the formula is referencing home team performance, a negative result means the home team should lose by that margin.  If I am playing a home underdog, it would be because the formulas are positive, or one is strongly positive and the other barely negative.  This happens alot, where you have a +1.25 and a -0.13, that is a case where I would play it, especially with a high ML, because what it is really telling me is: "on the one hand, I think the team will win by more than a run, and on the other, they will lose by a fraction of one, and so it seems relatively even, and yet the payoff is huge." 

If you could flip a dime and know that half the time you're going to get a quarter back and half the time you're going to lose the dime, wouldn't you do that all day every day?

3)  I thought i replied to you, but i guess not. So to answer your question from an earlier post, I mentioned that I knew the amount of games to bet before season starts. The reason is because I chase a team playing at home or away for 6 consecutive games.  Using data from the past 5 years, i look at how the teams perform on the road/home and look for any tendencies.  For example, certain teams in the past 5 seasons have always had at least 1 game on the road that went Under or loss.  I chase them to go under/ lose/win/over. 
I know that this is a very simple approach and is not derived from a statistical/mathematical logic. 

I'm sorry, but count me a skeptic.  A particular team's data from 5 years ago in a certain situation is hardly a predictive factor when personnel have changed so much and the environmental conditions are different, the motivating factors have changed, and so forth.  On the surface something like this can sound good, but when I've dug enough in the past on these kinds of things, I've always found flaws.  That's not to say it doesn't work for you, but I want to rule out anything I can't measure.  It's the only way I know how to manage money, whether in stocks or sports.  I'd rather be wrong and have a reason than right and lucky, because if I'm the latter I'm just a gambler.

4) I would like to try using your formula for the NHL this season and your same money management approach as well.
You mentioned that you've tried NHL last season, how did it go? and what formula did you use?

Would this be the formula:

(Home Goals per game + Road Goals against)/2 - (Road Goals per game + Home goals against)/2   ???

I do not know if it is possible to have a second formula for hockey because unlike baseball, you do not have the option of a "LISTED PITCHER".  Goalies are almost not taken into account for the lines as the backup can start last minute.
Also in hockey if you are looking for a 50% chance of winning....once the game gets into a shootout, it really is a crapshoot.

It's a different formula but based on similar factors; underdogs in favorable statistical positions with a attractive ML.  In hockey the goalie is important but obviously in a different way than the pitcher.  I don't consider shootouts because it's irrelevant; it's still a question of past performance and whoever scores the most wins, so regression formulas predict that adequately.  I'll share it as soon as I'm ready-last year was my first year on NHL and my records are a mess and I learned from starting the covers thread that I get overwhelmed with detailed questions I'm not ready to answer yet-I make one mistake and I'll get worn out for weeks.

I really like your Investment mentality in your method and i have followed way too many people in the past that just try to handicap the games and it is almost never profitable in the end. You have opened my eyes...with your ideas of efficiency and turnover rate.  And i now ask myself... WHY do i stop a chase at 3 games,,,or even 6 games? As there really is no statistical advantage other than the 'Home stand is over'.

Do ever stop to take a loss if you have a 10 game losing streak? or is 6 just happen to be the worst streak to date?

Thanks, I think you can see my philosophy is investment rather than gambling, and there needs to be a reason behind every decision.  But you've asked the one question I don't think I have a good answer for, and that is the chase.

I developed this modified chase because I thought most progressive systems were far too aggressive.  I also know that statistically speaking, in a 50/50 environment most losses will be immediately followed by a win, and that in a very high percentage of cases, if there are two losses in a row, a win is even more likely.  What I did not think through in depth (like my bankroll units), was the length of a streak.  So, at the moment six losses is the worst to date, 5 has happened once or twice, 4 several times, 3 often.  It's a perfect binomial graph, except I've never had 6 wins in a row (have had 5), so I guess I'm due.

The bottom line is I have to develop a method to address a 'endless' losing streak.  Statistically speaking, it is of course certain that eventually this will happen.  It is not a problem if you are straight betting, and I have considered eliminating the progressive system and just having more plays.  Thus far this season if I had bet twice a day instead of once a day, and eliminated the progression, I would have about the same profit, but have eliminated the progressive risk.  Of course, my efficiency would have dropped by 50%, so there is an element of risk there.

I know you are extremely busy, but i do hope to hear from you soon.

Hope this helps.

Another recent email exchange:

Hi, I have read through your baseball system posts on covers.com and I am very impressed. I have a couple questions and suggestions if you'd be willing to help me understand your methods. Questions: 
Thank you very much for your help. It is great how you have incorporated your experience in the business world into this successful system. Hope to hear from you soon.

Thanks for the feedback!

Mark

Surfer, where do you get your data?  I have KC at formula 1= -.18 and formula 2= -.14

Didn´t mean to sound like i´m complaining. Sorry about that. I just like to hear what people have to say about your system and maybe i can pick up a thing or two in the process. I don´t understand how people can mock you about a 20% increase per month. Since it´s virtually riskfree everyone should be so happy. The only reason i tried to modify the money management is that you talked about the possibility to bet two games per day and i think this involves less risk.
Input anyone?

Sdiinc where do you get your stats from I got formula 1 = .37 frm 2 = .915? I get my info from covers on the top of the game matchup page is a Home/Away page for R/9 and ERA. https://www.covers.com/pageLoader/pageLoader.aspx?page=/data/mlb/matchups/g4_homeaway_9.html

Also are you going to bet into the playoffs? If not are you planning on ending your system before the season ends? It would be good to end on a win rather than ending on a 5/6 game lose streak just a suggestion.

Forgot to say thanks for sharing the email conversations with us. A lot of good info there.

yea, it seems that there is a discrepancy in the stats.

washington looks like the best chance of an upset to me tonight

Thanks everyone for the feedback.  This thread is meant for that, after all.

I thought I had mentioned before that I am using a database I bought from a 'friend' last year.  And my formulas are being generated daily for my by a program another 'friend' put together for me last summer, rather than one I put together on Excel.  I don't want to alert my friend to the money I'm making because I know what will happen when I ask for his further input (i.e., suddenly the db is worth $1,000,000).  So, there is a little bit of the 'Wizard of Oz' going on here, but it's not intentional and I'm trying to work through it.

The reality of the matter is even though we're doing well I'm making more money at my real job and so I'm trying to balance this with my NFL picks and trying to iron out the NHL underdog formula and figuring out a NBA formula (barely made money last year).  No excuses, just a FYI.

So, bear with me, post your ideas/suggestions, and we'll all make (more) money together. 

Then everyone is obviously using different stats and or doing the formula wrong?

I think that the beauty of this system is the money management aspect.

Surfer, I think we all appreciate honest input.  However, like Milkman demonstrated for us before, not all criticism is constructive.

For example, you have not established as 'fact' that the formula I am using is different than that quoted above.  All you have done is assert that your input produces a different result than my input.  I am willing to bet that anyone who is following this thread will produce results different from others using the same formula; at least that is the feedback I have gotten by email over the last two months.

As I have already stated-multiple times-my formula is using a database I purchased, not a public or online source, and so there may be differences in the results produced by the formula.  This is of course the case with any formula that depends upon multiple inputs.

You chose to ignore this 'fact' in stating your conclusion, rather than accepting it as true and attempting to reconcile the differences for our mutual benefit.  You are confusing the facts for the truth-or you have another motive. 

You joined Covers this month and your only posts are on this page and they are all critical.  It would not be unreasonable for an objective observer to conclude your true intent is disruption. 

I welcome any post that is helpful.

Mark

CUB over MIL @ 110

However, I've got it MIL over CHC, but for the longer odds...I have STL over CIN,  DET over TEX and  CHW over NYY. Your picks are appreciated,  are you still adjusting your formula?

Phil

Mark, you seem like a very intelligent and scientific thinking man.so I think you could appreciate what surfer is trying to say, which is:

the ppg for home team, ppg for road team, era home...... ect
all the components of your formulas are not subjective. They are and can only be in this universe 1 value and 1 value only.

whether you get your info from espn, covers, mlb, sportsline, or a paid for database the bottom line is that the cubs played 148 games in which they scored 790 runs which equals 5.33 ppg. period.

this is his confusion.

surfer... I believe that  Mark is using a database  with several years of data.  Thats  where lies the discrepancy .  you  will not get the same  values  using  only  this  years  stats .

I think. 

so if each of you could respond with a simple yes if my thoughts in reference to each of you are correct,  it will be all cleared up and we can move on.



mark, we do appreciate your hard work. but men need to fully understand something before we can totally buy into it.  which is, I believe, what we all want to be able to do.