Jerry,

The only thing I am really interested in is finding out the relative standard deviations for TG and Rag.  What different results will \'mean\' is completely open to interpretation.

Based on the vast majority of posts on this board and Ragozin\'s board, the expected result is that TG will have a significantly smaller standard deviation than Ragozin for each race type -- the \'smoother line\' effect.  For my money, it is just an interesting fact to chase down -- just how big is the difference?  If, in fact, the difference is not terribly significant, then that would be something of an \'AHA\'.  However, there is nothing in the results that will help someone decide which product to buy, if that is what you are getting at.  I started to input the numbers today.  It will probably be another week before it\'s done.

As for you, Mr. Holbert, based on your posts seen this board, I would be very careful before engaging in a discussion of statistics.  If you have a point to make, make it and back it up with something.  Tossing out fact-poor -- or, in the case above, fact-less -- missives is just a big waste of cyberink and time.

Jerry,

I\'ll give you the same courtesy that you extended me in your last email.

Methodology:  I set up an Excel spreadsheet and input the names of each horse in two separate columns -- 1 for Rag. 1 for TG.  I will then subdivide each column into \'sprint\' \'route\' and \'turf\'.  I will then input the numbers from the respective brand of sheets into each spreadsheet until all numbers are entered and properly segmented.

Then, for each horse and each type of race, add the numbers and divide the total by the number of races, thus generating 1-3 data points for each horse (sprint, route, turf).

Next, I will take the average for sprint, route, and turft and subtract each number in the respective categories from the average, creating a \'deviation\' in the adjacent column.  Add the total, divide by number of races, and voila, another 1-3 data points for each horse.

Next, take all of the data points for sprint, route, and turf for both Rag. and TG, add them, divide by the number of horses in each sample size and I will have the average numbers for sprints, routes, and turf for each product as well as the standard deviation.  Then, compare TG to Rag.  The consensus is that TG numbers will be several points faster and that the standard deviation will be smaller.  Just how much is what I would like to know.

Anything else?

Now that the counterfeit ticket scam has been uncovered, any bets on when we will see calls for tracks to ban SAM machines?

Just an FYI from a Ragozin Sheets user.  I cashed two tickets today.

1) Starine to win in the F/M turf.  She was a huge overlay at 13:1 and had the second best numbers.

2) Trifecta in the Classic.  Granted, I fell into it -- big box tossing the bad pattern horses like WE, PD, and EA -- but Volponi was a solid semi-contender on the Sheets (not \'dog food,\' Scott) given that there were so many horses coming off ugly tops or who just weren\'t that fast.

Mall,

As a general rule (e.g., Thursdays at Aqueduct) you are right about where the majority of the handle comes from.  Two comments on that:

1) On the biggest racing days of the year (KY Derby, Preakness, Belmont, BC), the ones I really concentrate on (out of necessity more than anything else), the % of handle that comes from the \'unwashed\' is more than substantial.  I don\'t know the percentages but it has got to be over 50% of the mutuel pools.

2) Just because big bettors rule the parimutuel roost does not mean that they are all smart, sharp, or adopt the same handicapping theories about the \'bounce\'.  A lot of these guys believe in theories that are shaky at best and a lot of them are probably terrible money managers.  You can probably count on one hand the percentage of guys who are good enough to beat the takeout and as a % of total handle I\'ll bet it\'s well below 50%.

BTW, my standard deviation analysis project is stalled for now.  I have had no time to work on it for the past few weeks.  I\'ll get it done but it will be awhile.  Maybe by the X-mas holidays.

Mall,

By \'crowd\' I meant the betting public.

Mall,

I base my bounce predictions on the horse\'s established level, if there is one, and with an eye toward the absolute number it ran and whether horses of a similar age and gender have been able to maintain those numbers in the past.  Derby 1592 is right that 2 y.o. are the toughest to call but I\'m loathe to play any 2 y.o. not to bounce if they ran a 5 or better in their last race (Ragozin scale, of course).  That doesn\'t mean that they can\'t win but that kind of number is not the kind of figure off which you see many pair-ups or move-forwards.  

We\'ll find out eventually what Summer Colony, Sky Mesa, and Golden Apples ran but I can guarantee you that both SC and SM bounced.  Tougher call on GA.  Her race was pretty paceless and they came home in :22 and change so her fig may not look that good.

As it turns out, per my pre-race fears, there were not a lot of betting opportunities in the 10 stake races, or the other races on the three cards.  I didn\'t bet the Beldame but tossing SC and boxing the top two was a good bet.  The Champagne exacta was good and hittable if you were willing to eliminate the favorite -- which I was. The Ancient Title turned out to be a very good betting race.  Kalookan Queen at 9:2 was huge value and the exacta with Crafty C.T. paid a very healthy $41 for a $2 bet.

Even though I did not hit the 6th at SA it was very gratifying to see Congaree finish out of the money.

Thanks, Jerry.  This will take some time but I think it will be interesting.

Ragozin still has Pim 05/18/02 on their site and I downloaded it.  Can you make the card available on this site or tell me where I can find it?

Thanks.

Jerry,

What I am trying to measure mainly is not just the different scale but the difference in standard deviation -- the relative \'smoothness\' of the lines.  We hear so much talk about it and if we had a couple thousand data points then we could see about how much smoother TG lines are than Ragozin.

Actually, the Preakness day card was a great card to use b/c it had a ton of entries, lots of shippers, horses from NY and the Delaware Valley circuit which Ragozin has slower than you.  Unfortunately, I don\'t think that the Preakness card is still available on the Ragozin web site and I don\'t think I saved it to my hard drive.

I am aware of the other items you find interesting and they are, but everybody knows that you sometimes split the variant and Ragozin doesn\'t (with rare exceptions), so I\'m not sure we would learn a whole lot there that we don\'t already know.

I stand corrected.  I guess the weight difference was worth about 5 feet then.

Alydar,

I can add nothing to your latest exposition.  I agree.  

Some minor cleanup:

I was not trying to draw a direct analogy between handicapping and the stock market.  I was simply picking up on one common thread between the two \'markets\' -- the digestion and incorporation of readily available information into the price of an asset.

You still have not answered my question (I note HP\'s concurrence), which is fine b/c I don\'t think there is a handicapper alive with a positive ROI where the use of jockey ROI (not jockey skill, but Jockey ROI) is any meaningful part of his/her success.

When we talk about ROI, so-called \"hot streaks\" are relevant (and when I say \'relevant\' I don\'t mean relevant in that I want to incorporate it into my morning line but relevant in that you should not have so easily dismissed my point about \'hot streaks\') for two reasons: 1) a recent hot or bad streak will cause ROI to deviate from the long-term trend based on what you call \'established competence\' and 2) the crowd may overvalue recent success, often in part because of a recent surge in ROI.  My point, and I think we agree, is that once we have an opinion on the jockey, we factor that into the horses morning line.  His ROI, however, is not something to factor in.  I won\'t give horse \'A\' an addiitional 1% chance of winning on Thursday just because JD\'s ROI jumped from $1.60 to $1.85 in the last couple of days and neither would you.  Over time, of course, as jockeys improve or decline, we need to factor that in, but this exercise is at the margins.

Jerry, you may be saying this implicitly when you say that you \'look to see how the fractions came up versus the other races\' but I would think that you should look to how fast the field came home relative to the early fractions in the race.  For example, if the fractions of a race are:

24.3 49.3 114.0 137.2 149.0 it would seem more reasonable to consider a pace adjustment than if the race went:

24.3 49.3 114.0 138.2 149.4

The second race is slower but seems more likely due to a slow surface as opposed to a slow early pace.

Thoughts?

HP,

Glad to see that you got the money.  

I don\'t know why you want to bet into the hurricane that is Michael D.  What with all the ROI info he has and the work he has been doing getting insights into how Jerry Bailey feels when he wakes up in the morning (probably a lot like sausage making -- we don\'t want to watch that happening), you are cooked before the party even starts.  Quit while you are ahead.  Buy some T-bills or airline stocks.

BTW, I killed them at the Western Montana Fair.  I know my mules!

dp

Many a ruined day trader (and horseplayer) has followed this method, Michael.  God bless all of you.