Final Odds-They Don't Add Up?

[JustMissed]

I was doing a little test to compare a fair odds line as calculated by a certain handicapping program with the final odds as posted on the TSN chart. I could not get the chart odds line to add up to 100 for any race! I know ML odds are usually greater than 100 in order to consider racetrack take but I thought tote or final chart odds would always equal 100 to represent the amount of money bet to win on each horse compared to the total win bets.

Where am I going wrong? Please let me know cause I could be losing my mind.;)

Thanks,

JustMissed

[anotherdave]

If the track odds they added up to 100 we'd all be rich!

The take is in the final odds, so they will add up to more (say around 125% depending on track take)

AD

[JustMissed]

That makes sense. I thought it was just the morning line odds that included the take.

My little test of fair odds did not work out anyway. I have concluded that they are nothing more than a mathematical expression of someones opinion of how a horse should run. There is really nothing to compare them to except the final order of finish.

Thanks for your reply,

JustMissed
:)

[Rick]

JustMissed,

If your method combined with actual odds can predict finish position significantly better than actual odds alone, then you have a winning method.

[JustMissed]

Very well put.

My little test was to prove to myself that fair odds lines are not verifiable but are only a "predictor tool" and a "betting tool".

In reality, when you make up a fair odds lines you are only listing the horses in the order that you think they will finish.

I have given up on making use of fair odds lines because you have to balance it out to 100, which means you need to handicap each horse in the field. I am a throw out capper and can usually eliminate almost half the field in a few minutes and don't have the time to waste assigning odds to horses I'm not going to play anyway.

The only other choice is to let the computer assign the fair odds, but if I can't look at the program code to figure out what it is using I don't have any confidence in someone else numbers anyway.

I am learning very quickly to be skeptical of everything I read or hear.

Just my take,

Justmissed


:)

[Rick]

JustMissed,

You don't really need to calculate odds. If your top selection is 3rd favorite or higher, you probably have an overlay. Even a 2nd favorite may be an overlay if you have a very good method.

[ranchwest]

So a favorite cannot be an overlay?

[anotherdave]

I don't get that feeling from Rick's post. If my top choice is 3rd choice+ with the bettors I would be pretty sure it was an overlay. If my first choice was the favourite it could be an overlay, but not nearly as often.

Putting it simply:

If my first choice was 10-1 and 5th favourite, I don't really have to do much but bet. I wouldn't think twice.

If he's 9-5 favourite, I might still bet, but not always.

I even bet the odd 1-1 shot that is my first choice, but I'd have to feel pretty sure about him.

If he's 2-5, I'll wait for the next race.

AD

[Rick]

ranchwest,

I don't mean to imply that a favorite can't be an overlay, but you'd have to compute odds in order to find it in that case. There are a lot fewer legitimate overlays among favorites though and they tend to be less profitable overall. So, you're not giving up too much by just ignoring them and focusing on the longer odds horses which are both more lucrative and easier to identify.

Another thing you could do is look a how much difference there is between you top rated and second rated horses and set up some rules for minimum odds. That would allow you to select favorites some of the time.

My procedure is to use morning line odds and always throw out the morning line favorite, mainly because it's much easier than monitoring actual odds and works about as well. So I wind up on favorites sometimes if they were bet down from their morning line rank. That's a positive thing more often than not, at least with my method.

[TenZin]

I don't understand why you spend so much time with
Over & Under Lays? Just bet your top selection to Win or Pass
the Win Betting; Bet your Top 2 or 3 in some Exacta combination.
If your system is a good one you will win money after X number
of races. If not, take up bowling.

[anotherdave]

Different things work for different people. There are no absolutes. I tried basically your technique and it didn't work for me - might for you, but not for my selections. When I started using the overlay concept I started winning. It would be hard to go back.

AD

[Rick]

Why do people get so angry when anyone mentions overlays? An overlay is a horse that wins at a higher rate than it should according to it's odds. If your top selection accomplishes that in every race, you should soon be a millionaire. More likely, you'll have some set of rules for eliminating underlays and won't have a play in every race. I hardly ever monitor the toteboard on a race and have been getting overlays for many years. It's true that a few people like to calculate their own odds and compare them with the actual odds, but it's usually not necessary.

I've heard people at the track and ignorant TV analysts refer to horses with actual odds greater than morning line odds as "overlays" and maybe that's where the confusion comes from.

[David McKenzie]

Originally posted by Rick
Why do people get so angry when anyone mentions overlays?

Rick,

This is germane in the sense that my friend and I had a conversation that turned into a serious disagreement originating with 'overlays.'

This is a copy of a letter I sent to HDW last night. I haven't heard from them yet. You don't perchance happen to have this data easily available, do you?

Thanks,
David

// I'm a subscriber to your service and am very happy with everything.

Tonight my friend and I had a heated discussion on the way home from the track.

I said horses that go off below their morning lines odds have a higher Win % than horses that go off above their morning line odds. My buddy vehemently disagreed.

Who's correct?

Is there someplace I can scrounge up data for this, or would you please be kind enough to send me stats, if it's not a hassle for you?

Many thanks,
David McKenzie //

[ranchwest]

Originally posted by Rick
Why do people get so angry when anyone mentions overlays? An overlay is a horse that wins at a higher rate than it should according to it's odds. If your top selection accomplishes that in every race, you should soon be a millionaire. More likely, you'll have some set of rules for eliminating underlays and won't have a play in every race. I hardly ever monitor the toteboard on a race and have been getting overlays for many years. It's true that a few people like to calculate their own odds and compare them with the actual odds, but it's usually not necessary.

I've heard people at the track and ignorant TV analysts refer to horses with actual odds greater than morning line odds as "overlays" and maybe that's where the confusion comes from.

I don't get angry, but I do get confused because people don't recognize what you've pointed out. Comparing to the tote board is usually not necessary.

If you know your methodology returns a profit and you have a good idea of what most horses are going to pay, why is the tote board mandatory?

[Rick]

David,

I don't think that horses who go off at lower than morning line odds have a higher win % BUT they do have a better ROI. Here is an example from a Sport Stat study from 1995:

Favorites Overall: 32.8% wins, -0.20 ROI.

Post Time Favorite and ML Favorite: 34.5% wins, -0.22 ROI.

Post Time Favorite, not ML Favorite: 29.9% wins, -0.16 ROI.

That 6% difference, to me, is a very significant difference.


ranchwest,

I can't tell you how many otherwise intelligent people I've run into that thought that they could easily throw together a regression model to predict odds and make a million betting "overlays". As I've pointed out before, unless your method adds significant information AFTER actual odds is added to your model, you'll LOSE betting what you think are "overlays". It's a hard lesson to learn for those with a little bit of knowledge about statistical methods.

[Derek2U]

One very interesting comment came from Dave when he said
that "figuring over-under-lays made him a better selector".
Was that only confined to higher odds gotten or did it also make
you a more discerning selector? I suppose whatever works is
great, but I also think the BEST system is one having a VERY HIGH
% of winners among its TOP TWO picks. From 50-70 % would
be quite enough to Win BIG -- and consistently also.

[tanda]

"One very interesting comment came from Dave when he said
that "figuring over-under-lays made him a better selector".
Was that only confined to higher odds gotten or did it also make
you a more discerning selector? I suppose whatever works is
great, but I also think the BEST system is one having a VERY HIGH
% of winners among its TOP TWO picks. From 50-70 % would
be quite enough to Win BIG -- and consistently also." From Derek2U

Not necessarily. Betting the top two public choices wins around 53% but obviously throws a loss.