Vegas vs KenPom Week 8

Going back to last week, your listed Offical Plays that were Road Dogs are 19-8-1 ATS (again, counting Rider a push). 67%.  Also, 13 of those 19 wins were outright winners, with quite a few double digit dogs.  So, 13 of 28 games won outright.  

Today, 5 RDs covered, 4 won outright, one was +15 or at least +925 ML.  Based on these numbers, it seems to hold great value to pick out all the RDs and play them on the ML.  Today, a $100 player would have netted $1300 playing just the ML on RDs. 

Even at an average of +300 ML - +400 ML, the past week and a half would have netted a $100 player $2400 - $3700, as opposed to $1020 laying $110 to win $100 taking the points in each game represented with a RD. 

Perhaps a greater success rate, and fewer plays, would be to create your full list of official plays and only play the RDs, maybe playing the ML and taking points.  What do you think?

Flyer, I know that beaver has pointed out the better success for Away picks, but I didn't notice a huge difference between RDT and RDA, that's why I think I'm going to try the ML on all RDs.  I could be wrong, though, and the sample size I took was only the past 11 days.

Well, all the discussions I've had over the years, up to this point & in this thread, have been "academic".

We knew we were all "putting the cart before the horse" when we were discussing "money mangement" outside the realm of the predicate, "positive expectation"

The best we could do with "innate" handicapping skills AS OPPOSED TO "analytical" skills which were arbitrarily derived was to "guess" at our edge.

The thing that everyone needed in the first place was something that analytically(power ratings???) produced a "quantifiable edge" that could be measured against a "fair coin test" so familiar to students of "beginning probability" classes.

If the jury is still out on this topic, then the jury is not far from reaching the decision that indeed, beaverfan23 has found something which produces something better than 50/50 with 99% certitude.

167 wins - 134 losses using beaverfan's stated results

Since our bankrolls are all MONEY THAT WE CAN AFFORD TO LOSE then we can deal with less certitude; to wit, we can be 75% certain that 167 wins in 304 wagering events (54.9% winning proficiency) @ -105 gives us AT WORST an "unfair coin" that is 51.4% unfair

...we can then be 75% certain that our optimal wager @ -105 would be 51.4 - (48.6/.95) = 0.24% or 1/4 of 1% which would necessitate approximately 400 units of bankroll.

(parenthetically...at 75% certitude, we may also be as high as 58.3% proficient by hitting 54.9% over 304 wagering events)

so let's fire off 86 games all going off simultaneously and we need 20% more bankroll for 480 units. Let's round that up to 500 units

5Dimes offers as little as $0.50 per wager  Internet betting @ -105 and they will also cover the Western Union fee + will cover the payout fee if done on some certain day of the week for a minimum $250 deposit (500+ units)

So we can be 75% certain that we are NOT OVERBETTING by wagering 1/500th of our bankroll on each wager and each simultaneous wagers (15 games going off all at once).

...and this is by betting on EACH & EVERY game that beaverfan23 has filtered out for us.

This is why I encouraged beaverfan23 to get as manybets/wagers with a 54%  basement as possible; to wit, ..."we need more input  data" for the "unfair coin" test.

So 1/4 of 1 % is the WORST-CASE SCENARIO for bettinng safely.

That is why I like Bobcat36's betting scheme based on filtered data stated in an earlier post.

1 betting/wagering unit of 1.5%( divide bankroll into 70 equal wagering units) is probably still safe with a 65% (17 out of 20) certitude over the 304 wagering units and betting all 304 games.

This is obviously a "labor of love" for beaverfan23... meaning that he has put in a considerable amount of time and expended a considerable amount of energy to share this with us.... the fruit of this is that he has had over 8000 views to this thread.

I therefore would like to take the opportunity to thank him now for extending himself on our behalf.

Sharpmoney:

Those are the only options for making single wagers when you have 2 variables (home and away) and (fave vs. dog) and one constant (line moves away).

Ishmael:

WOW Could you break that down for a journalism major. To much math.

Also, GREAT idea on the ML for road dogs.

Certainly Bobcat36,

Let's pretend Beavefan23 is flipping a KenPom two-headed coin which he is using to predict the results of basketball games at a -105 5Dimes betting line.

But the KenPom coin does not predict perfectly; at present it is predicting 54.9% over 304 wagers.

What we want to know regarding that 54.9% figure is this?

1) Is beaverfan23's KenPom coin just getting "lucky" and it does not truly represent 54.9% proficiency.

Answer: We can be certain that if beaverfan23 flips that coin 304 times each time in 4 flipping sessions that he is no worse than a 51.4% bettor... meaning we have a goose that is laying a golden egg since that gives us an edge of 1/4 of 1%

...and we would expect that figure to be reliable 3 out of the 4 flipping sessions.

We're hoping here that the true value/expectation lies in the 3 sessions and not the fourth.

2) But what IF INSTEAD of being "lucky" with those 304 coin flips and only working to a true edge of 51.4%, what if beaverfan23 has been UNLUCKY in picking only 54.9% winners?

...well, we can be 75% certain that beaverfan23's KenPom coin could TRULY be a Cadillac Goose that is laying Golden Eggs and that the "bad luck" he has been having in only picking 54.9% winners MASKS THE FACT that he is truly a 58.3% handicapper and just has been having a string of "bad luck" which has resulted in only 54.9% winners.

This is why your "money management" proposal was both balanced and well-considered, Bobcat36... it strived to take advantage of the POSITIVE EXPECTATION yet add a measure of safety.

The key is this... even if beaverfan23's kenPom coin is having the worst streak of luck possible, we can be somewhat secure in the knowledge that we still have a goose that is laying golden eggs

Quote Originally Posted by beaverfan23:

Kahlmy

I'm attempting to add you as a covers.com/wagerline friend... there may be a "friend request" in your inbox.