Locking Cubs +102
GLTA!!!
GLTA!!!
-
Comment -
Updating....
YTD 3-1
Inital Bakrol $1000.00
Balance $1042.42
At risk $43.43
open plays... Cubs 102Comment -
No play on White Sox!!Comment -
locking Rays +115
GLTA!!!Comment -
Updating....
Initial bankrol $1000.00
Balance $1000.73
At risk $85.12
Open plays cubs+102 $43.33, rays+115 $41.69.
GLTA!!!Comment -
Updating....
YTD 4-1
Initial bankrol $1000.00
Balance $1088.25
At risk $41.69
Open bets Rays +115Comment -
Updating.....
5-1 YTD
Initial bankrol $1000.00
Balance $1177.88
At risk $0
No open plays
good nite and GLTA!!!Comment -
Good show....rather outstanding!
Comment -
Let's goooooComment -
Games for today that might qualify.... Phillies, Brewers and Braves, will be back later.
GLTA!!Comment -
Cubs might become a play,..... so for now model has 4 games that might qualify, Phillies, Braves, Brewers and Cubs... will keep updatingComment -
I know lol!!.. I try to just follow my model, sometimes I scractch my head with the plays that come up, but hey had worked for 7 years so I just wait and if a bad team qualify so be it I´ll play it cheers!!!Comment -
locking up Phillies +118 $47.11
GLTA!!Comment -
locking up Cubs +108 $45.23
GLTA!!!Comment -
No play on O´s.....
GLTA!!!Comment -
locking up Braves +114... $43.42Comment -
locking up brewers +120 ...$41.68
GLTA!!!Comment -
Updating.....
YTD 5-1
Initial bankrol $1000.00
Balance $1000.44
At risk $137.44
Open plays Phillies +118 $47.11, Cubs +108 $45.23, Braves +114 $43.42, Brewers +120 $41.68
GLTA!!!!Comment -
Interesting thread you have going here, thank you for sharing! I have a few different models I've been experimenting with myself (been focusing most of time and effort on developing a stronger underdog system), and it seems to spit out many of the same plays as your model, which of course gives me hope it will be successful!
Just had a quick question regarding your money management. The fact that you use the Kelly management system of course shows you know how to maximize your profits, but is there a particular reason you re-size your bankroll before each individual wager? I'm assuming it's because it is unknown how many plays the model may have throughout the day (based on line movements, etc.) - one hour it may fit into the criteria and the next it falls out? The only reason I ask is because I've always wondered if that was the most effective method of utilizing Kelly, and I've never actually experimented with it myself. Then again, the models I utilize already know (for the most part) how many plays it will have, with the exception of one or two later games added on depending on line movement and other filtered variables. That's just always been something I've wanted to research but have never found the time to do, resizing roll after each wager vs. betting set 4% (and exact same amount) on each play based on BR size at start of day. Also, how do you decide what order they are wagered - for example, was the Phillies, Cubs, Braves, then Brewers order that particular way because that's the relative strength of each play as rated by the model?
Sorry for the long post and load of questions, just very excited to follow along and discuss various systems/stats/models to help each other enhance current ones and develop new ones! Much of the knowledge and methods I utilize are adopted from "Conquering Risk" by Elihu Feustel and George Howard (very good book for anyone interested in developing their own models).
BOL on these and the rest of your plays tonight and this weekend, my friend! Looking forward to discussing and cashing plays throughout the season!
Comment -
First of all thanks for stoping by, well let see if I can explain your questions in an understandable way, lol.Interesting thread you have going here, thank you for sharing! I have a few different models I've been experimenting with myself (been focusing most of time and effort on developing a stronger underdog system), and it seems to spit out many of the same plays as your model, which of course gives me hope it will be successful!
Just had a quick question regarding your money management. The fact that you use the Kelly management system of course shows you know how to maximize your profits, but is there a particular reason you re-size your bankroll before each individual wager? I'm assuming it's because it is unknown how many plays the model may have throughout the day (based on line movements, etc.) - one hour it may fit into the criteria and the next it falls out? The only reason I ask is because I've always wondered if that was the most effective method of utilizing Kelly, and I've never actually experimented with it myself. Then again, the models I utilize already know (for the most part) how many plays it will have, with the exception of one or two later games added on depending on line movement and other filtered variables. That's just always been something I've wanted to research but have never found the time to do, resizing roll after each wager vs. betting set 4% (and exact same amount) on each play based on BR size at start of day. Also, how do you decide what order they are wagered - for example, was the Phillies, Cubs, Braves, then Brewers order that particular way because that's the relative strength of each play as rated by the model?
Sorry for the long post and load of questions, just very excited to follow along and discuss various systems/stats/models to help each other enhance current ones and develop new ones! Much of the knowledge and methods I utilize are adopted from "Conquering Risk" by Elihu Feustel and George Howard (very good book for anyone interested in developing their own models).
BOL on these and the rest of your plays tonight and this weekend, my friend! Looking forward to discussing and cashing plays throughout the season!
Regarding the re-sizing of every single bet you are correct I decided to do it that way due to the lack of insurance on how many games will qualify any given day. some may open outside the model criteria and some may open inside , on the backtesting I made for the model (7 years) I found days with as many as 5 plays and many days with none, I made that backtesting using flag betting, Kelly in the range of 1% up to 5% and since the model is built in order to doublé up the initial bankrol the fastest way without risking too much was 4%. The order of the plays are made by the first one to reach the criteria and "stays" for the period of time I set in the model, it´s the first one to be made if any. As you can notice the model rely some of the criteria on line movement.
I have tried many variables and after using from situational capping, stats and algorythms, flip coins ...hahaha just kidding, I came with a grinding model that has doublé up the initial bankrol in an average of 200 plays.
Indeed you need a lot of time to do simulations with your model regarding bankrol management. I am sure you are on the right path. Looking forward to share and learn.
Cheers!!!Comment -
Something you may ask.... and if not anyway I would like to add ...lol.
As an example 2014 mlb seaon , there were cold streaks up to 6 losing games or 2-8, and then with out tweaking anything came up with streaks o 5 or 6 in a row or 12-3 and so on, why did I decide to set the goal to doublé up the bankrol? At the begginig I thougth it was a bad model, and was just luck, well as in ´14 there were similar outcomes in ´13, ´12, ´11, ´10, ´09, ´08. I spent lot of time backesting and came up with sounding results, as I mentioned before with an average of 200 plays ever single year inital bankrol was doubled up, funny thing is that when reaches it´s peak will start to come down at máximum speed, and when book´s (this is just a paranoia conspiracy made by myself lmao) get most of their money back the model starts to give winners at a high rate, at the end of the season 450 plays avg model have come u with a doublé up bankrol. Luck, fact , fiction? I don´t know but numbers don´t lie and I would take any model that doublé up my bankrol.
Comment -
Updating....
6-4 YTD
Initial Bankrol $1000.00
Balance $1092.14
At risk $0
No open plays
GLTA!!!Comment -
That's awesome man, thank you for replying, with such details too! Yeah the main downside/problem I have with my models is that I try to test different Kelly styles to find the most effective way to maximize profits, and it seems that whenever I start trying more aggressive approaches (betting higher % of bankroll bc of a miscalculation/overstatement of my perceived 'edge'), it just so happens to line up with the exact time the model hits a losing streak, thereby eating away chunks of my bankroll! Like you, I always seem to think it's more than coincidence when they always happen to line up at the same exact time! Me - 'ok, this is definitely profitable, so I can try a more aggressive Kelly approach here, I would have made X amount more had I wagers it this way' - 2 weeks later 'hello worst losing streak the model has seen to date, what impeccable timing to make an appearance!'
What is the winning % of your model plays? The book I mentioned earlier (Conquering Risk) has an awesome formula to help determine what % of bankroll one should invest based on average odds and actual winning % of the picks. The most amazing aspect of betting on baseball dogs (and a stat very few people actually know) is the fact that overall (all dogs over a given season at + money regardless of odds) they actually win at a fairly consistent rate of around 40% or so. Therefore, all one really needs to do in developing their underdog model is find an edge that increases that to 50% of the time, bc that is all you really need to make a decent profit when the break-even % is 50% or lower! That's the true beauty of wagering on dogs, being able to win at a smaller rate and still make huge profits (the higher the winning %, the higher the profits of course).
Of course you've already researched and discovered that 4% works best for your specific goal of doubling your bankroll, but have your tried backtesting with a different agenda/goal other than doubling? Like calculating anything within your range (or even higher than 4%) for an entire season to see what the results would be?
This is very fun actually getting to discuss stats/sabermetrics/developing models with other people who actually do the same exact thing! It's nice actually knowing I'm not the only math 'nerd' and numbers guy out there crunching numbers and testing to improve or create models! Lol
I know this season will be amazing and fun knowing that we'll all be rolling in the money by the time fall comes around!
Comment -
The winning % year to year have been 49.1% the lowest and 50.3% the highest, about the book I read it already pretty nice book btw. After Reading yoour comments and as I am always looking to get better, some ideas have come to mind, I will throw all the outcomes on some Elliot waves pronciples and will try to find some patterns in order to try to identify where the cold streaks happen and so use a different approach on the % of the bankrol to be invested or none.
I´ll give you an example:
2014
First wave (let´s call it that a wave of plays) 99-96 , 50.7% bank rol from 1000 went up to 2065.11
second wave 13-25 , 34.21% those 2065 went down up to 1300 and some change
third wave 11-4 73.33% from 1300 to 2100 and change
fourth wave 3-9 25% from 2100 to 1500 and change
fifth wave 10-1 90.91% from 1500 to 2400 and change (highest amount during the season)
sixth wave 3-15 16.67% from 2400 to 1500
seventh wave 11-5 68.75% 1500 to 2100
eighth wave 10-17 37.04% 2100 to 1600
ninth wave 12-6 66.67% 1600 to 2200
last wave 7-12 36.84% 2200 to 1897 and end of the regular season
And so the rest of the seasons I have backtested thow similar results, that is the reason I set the goal of doubling the initial bankrol and call it a season. I believe this model could work better and maximize the profits in the long haul, I see it as long term invesment and IMO it´s the best way to be profitable on any investment. Might not be fancy to aim in a long term profit but I believe on it. As bigger get your bankrol bigger will be the benefit. Very simple If you want to profit a million you must invest at least another million, lottery tickets are cheap but hey when was the last time you hit it?
Comment -
I like this....you guys are providing an education...
Comment -
I just finish to throw all '14 season on some elliot model.... outcome is pretty intresting, I have narrowed the criteria on my model looking to eliminate the constant on the cold streaks and '14 season went from a 179-190 record 48.5% to a better 94-91 50.81% with a 9.59% ROI, hey even I can set the goal now to make 1000 bankrol into a 3000 one. Will backtest from ´08 to ´13 and see what comes up.
Comment
