2011 Racetrack Rankings and Resource Released

[DeanT]

For a web copy of this release, please visit here:

http://blog.horseplayersassociation.org/2011/07/2011-hana-track-ratings-released-here.html

2011 HANA Track Ratings Released: Keeneland Tops List for Third Year Running, Monmouth “Elite Meet” and Retama Move Higher, California Tracks Fall

FOR IMMEDIATE RELEASE

(Charlottesville, VA, July 7th 2011): The Horseplayers Association of North America has released their annual Track Ratings. In 2011, Keeneland, which is currently celebrating its 75th anniversary, was the number one rated track in North America.

“With takeout rates of no higher than 19% on any bet, a field size of over 9 horses per race, and an average pool size of over $700,000 per race, Keeneland came out on top once again.” said HANA President Jeff Platt.

“We are so honored for Keeneland to earn a third consecutive ranking as HANA’s top racetrack in North America,” Keeneland President & CEO Nick Nicholson said. “Our great team at Keeneland strives every day to go the extra mile in delivering quality service to our fans, and I believe that is what sets us apart. I sincerely thank all the wonderful Keeneland employees for their efforts in achieving this milestone.”

The HANA Track Ratings are based on an algorithm designed by HANA Board member Bill Weaver, a retired engineer. Using studies and empirical data which are directly correlated to horseplayer value and handle growth, namely takeout rate, field size, wager variety, pool size and signal distribution, a composite score is tabulated, and tracks are ranked. The rankings are posted on HANA’s web page, which allow horseplayers and potential customers to sort data on the various categories, as a resource.

This past year there have been several changes in the way racetracks have catered to their customer base, as some have lowered takeout and/or concentrated on providing a bettable product via an increase in field size. These were reflected in the algorithm, and its resulting rankings. The wildly popular “elite meet” at Monmouth last season vaulted the oval from 34th, to 6th. Retama, who has been struggling without alternative gaming, has also tried its best to cater to customers. Their low 12% takeout on doubles and pick 3’s as well as their huge field size of 10.62 horses per race (tops in North America) moved them from 10th to 4th.

Third ranked Tampa and fifth ranked Gulfstream also scored well and have the most potential to move up and challenge Keeneland’s number one ranking in the future. Their high three and four horse vertical wager takeouts are weighing them down somewhat in the rankings.

On the opposite side of the spectrum, the historically highly-ranked California tracks took a tumble. With some of the highest two horse exotic takeout rates in North America, Del Mar (8th in 2010, 26th in 2011), Santa Anita (13th in 2010, 37th in 2011), Hollywood Park (15th in 2010, 30th in 2011) and Golden Gate Fields (32nd in 2010, 48th in 2011) all lost ground.

The complete list of 2011 racetracks and all available sortable metrics can be found at: http://www.horseplayersassociation.org/hanatrackratingsbyoverallscore2011.html

To look at our methodology and algorithm for the ratings system, please access the player resource section of our website here : http://www.horseplayersassociation.org

To contact us at HANA, please email us at info @ hanaweb.org.

For a web copy of this release, please visit here:

The Horseplayers Association of North America is a grassroots group of horseplayers, not affiliated with any organization, who are not pleased with the direction the game has taken. HANA believes that both tracks and horseman groups have become bogged down with industry infighting and have completely forgotten something: The importance of the customer. HANA hopes, through proactive change on several key issues (including but not limited to), open signal access, lower effective takeouts, wagering integrity, affordable data and customer appreciation, the industry’s handle losses can be reversed. HANA is currently made up of over 2000 horseplayers (both harness and thoroughbred) from almost all states and Canadian provinces. It currently represents over $75,000,000 of yearly racing handle.

Our web address is http://www.horseplayersassociation.org and interested horseplayers can sign up there for free. We are horseplayers, just like you and we are trying to make a difference. We need, appreciate, and ask humbly for your support.

[andymays]

Outstanding work by everyone involved! :ThmbUp:

[OTM Al]

Always interesting to look at such reports. The first question that comes to mind is what is the time period over which the data pertains? The press release talks about Monmouth getting propped up because of the "Elite Meet" last year, though this year is an entirely different situation. But at the same time California is being penalized for this year's takeout raises. I couldn't find the answer to this in the methodology. Without an explanation, this looks like cherry picking to reward friends and penalize enemies, which then calls the validity of the study into question. I can understand wanting the most up to date info in this, as it best serves it's purpose by being timely, but using Monmouth as a highlight seems to be doing just the opposite because ithe example given is not current.

[chickenhead]

Believe they have always used current takeout rates (as current as possible) and previous period field avg. Obviously with meets beginning and ending at different times, and takeout rates and field size changing all the time, there is never a static environment that lines up across the board. Since takeout is known concurrent with a meet beginning, and avg field size only know after the period in question has concluded, it seems like a logical, if necessarily imperfect, way to do things.

[OTM Al]

Believe they have always used current takeout rates (as current as possible) and previous period field avg. Obviously with meets beginning and ending at different times, and takeout rates and field size changing all the time, there is never a static environment that lines up across the board. Since takeout is known concurrent with a meet beginning, and avg field size only know after the period in question has concluded, it seems like a logical, if necessarily imperfect, way to do things.

Kinda figured it was like that, but then in that case, it would probably much more appropriate to use 2009 field size and purse data for Monmouth rather than 2010 as 2010 is clearly an anomaly and, as I am an example, citing Monmouth prominantly makes the situation very confusing as to the vintage of the data being used and can, using it along side what is clearly this year's CA takeout data, make the report look biased. Any such implications need to be avoided on things like this as they can hurt credibility when there is no reason that should be.

[Charli125]

Kinda figured it was like that, but then in that case, it would probably much more appropriate to use 2009 field size and purse data for Monmouth rather than 2010 as 2010 is clearly an anomaly and, as I am an example, citing Monmouth prominantly makes the situation very confusing as to the vintage of the data being used and can, using it along side what is clearly this year's CA takeout data, make the report look biased. Any such implications need to be avoided on things like this as they can hurt credibility when there is no reason that should be.

Chickenhead called it right. Field size and Pool size use the previous year totals(in this case 2010), and takeout uses as current as possible. As the year progresses, takeout changes are listed as updates, but are not included in the aggregate rating because that would require recalculating the whole ranking.

Obviously things are constantly changing, so we want to keep the resource current for players wanting to check takeout, but it's just not reasonable to update the aggregate ratings, field size, and pool size throughout the year.

And regarding Monmouth, even though it's an anomaly, we wouldn't want to treat any track different from the others. That's why we used 2010 data for all tracks.

[OTM Al]

Chickenhead called it right. Field size and Pool size use the previous year totals(in this case 2010), and takeout uses as current as possible. As the year progresses, takeout changes are listed as updates, but are not included in the aggregate rating because that would require recalculating the whole ranking.

Obviously things are constantly changing, so we want to keep the resource current for players wanting to check takeout, but it's just not reasonable to update the aggregate ratings, field size, and pool size throughout the year.

And regarding Monmouth, even though it's an anomaly, we wouldn't want to treat any track different from the others. That's why we used 2010 data for all tracks.

That's absolutely fine technique, though it should be clearly stated. However, it still seems to me that in a case like Monmouth, in which you know last year's numbers are skewed, other numbers should be used. Normally last year's numbers should be an unbiased estimate of this year's condition, but in Monmouth's case, that just isn't true.

Don't want anyone to take this as personal. Used to have to read econ papers and make such comments, so what I'm saying is a purely academic criticism. Thus to me it is wrong to trumpet Monmouth based on numbers we know do not represent current realities. 2009 would be unbiased, or at least a lot less biased, estimates than 2010. And as long as that is noted, it is much more legitimate than just using 2010 because you did that for everyone.

[Horseplayersbet.com]

Kinda figured it was like that, but then in that case, it would probably much more appropriate to use 2009 field size and purse data for Monmouth rather than 2010 as 2010 is clearly an anomaly and, as I am an example, citing Monmouth prominantly makes the situation very confusing as to the vintage of the data being used and can, using it along side what is clearly this year's CA takeout data, make the report look biased. Any such implications need to be avoided on things like this as they can hurt credibility when there is no reason that should be.
There is no perfect solution here. I think the majority of people are OK with 2010 betting data and current takeout info. Old takeout from last year still shows.
Lots of things have changed this year: A few tracks cut back a day a week and handle is down another 8% on the year or so.
The data here serves its purpose and I believe it is a very handy resource for most people.

[chickenhead]

Kinda figured it was like that, but then in that case, it would probably much more appropriate to use 2009 field size and purse data for Monmouth rather than 2010 as 2010 is clearly an anomaly and, as I am an example, citing Monmouth prominantly makes the situation very confusing as to the vintage of the data being used and can, using it along side what is clearly this year's CA takeout data, make the report look biased. Any such implications need to be avoided on things like this as they can hurt credibility when there is no reason that should be.

Don't know, I kind of think not changing the rules or calculations makes the ratings seem less biased, they might be less than great for a forward look at monmouth, but biased certainly isn't the right word. Same goes for California, they were ranked highly before, for which Hana was sometimes accused of bias, now they are ranked lower, which other people will probably consider a bias. But the methodology hasn't changed.

[Charli125]

Don't want anyone to take this as personal. Used to have to read econ papers and make such comments, so what I'm saying is a purely academic criticism. Thus to me it is wrong to trumpet Monmouth based on numbers we know do not represent current realities. 2009 would be unbiased, or at least a lot less biased, estimates than 2010. And as long as that is noted, it is much more legitimate than just using 2010 because you did that for everyone.

Thanks for the input Al, and we won't take it personally! Feedback is very important and the Track Ratings change a bit each year based on just this type of feedback.

I think the point of the press release portion about Monmouth was to explain why they made such a jump. Without giving any explanation it would leave the uneducated viewer thinking that the numbers shown were indicative of future numbers.

It's not perfect, but I do think we're pretty balanced in terms of making it a useful resource for the players, while maintaining the rating system.

[OTM Al]

Don't know, I kind of think not changing the rules or calculations makes the ratings seem less biased, they might be less than great for a forward look at monmouth, but biased certainly isn't the right word. Same goes for California, they were ranked highly before, for which Hana was sometimes accused of bias, now they are ranked lower, which other people will probably consider a bias. But the methodology hasn't changed.

I mean biased as in statistically biased not opinion based biased. Clearly Monmouth is a biased result in this poll. It goes up 6 spots because of the "Elite Meet" which is a situation we know no longer exists and now we know they are much more like 2009 that 2010 in terms of those factors this year. If you want to use 2010 across the board, fine, but in that case I'd put a big "*" on Monmouth and certainly not use it as a part of my abstract/press release.

[chickenhead]

ok, you mentioned it appeared Hana may be picking and choosing winners due to lack of clarity, which isn't statistical bias, that is the bias I was speaking to. But regards to monmouth- I would still think no. The track ratings aren't a prediction of the next years track ratings, they are meant to be closer to a measurement. If the data is noisy, its noisy, but I think sticking with the actual data is a good idea. Replacing data with estimates or massaging the method to get a specific result, even if it "seems better" is something I would advise Hana against. If credibility is a concern, and it is, that box simply shouldn't be opened. Stale results in some cases are 100x preferable to managing the outcome, good intentioned or not - even if it leads to "better" results.

They can leave it up to the readers to apply there own secondary analysis (like that x is an "outlier).

[DeanT]

I kinda like the MTH stats because it shows what a track needs to do to grow handle. MTH handle was up huge with close to a half billion bet last season. If other tracks try to do what they did, it can make a big splash, help horseplayers and help racing.

Horseplayers know the meet has changed this year, and those who don't were told in the PR specifically it was the elite meet numbers. Most horseplayers are pretty sharp.

[OTM Al]

ok, you mentioned it appeared Hana may be picking and choosing winners due to lack of clarity, which isn't statistical bias, that is the bias I was speaking to. But regards to monmouth- I would still think no. The track ratings aren't a prediction of the next years track ratings, they are meant to be closer to a measurement. If the data is noisy, its noisy, but I think sticking with the actual data is a good idea. Replacing data with estimates or massaging the method to get a specific result, even if it "seems better" is something I would advise Hana against. If credibility is a concern, and it is, that box simply shouldn't be opened. Stale results in some cases are 100x preferable to managing the outcome, good intentioned or not - even if it leads to "better" results.

They can leave it up to the readers to apply there own secondary analysis (like that x is an "outlier).

These of course are the exact sort of arguments that always arise over such matters and both sides always have valid points. The only thing that all can generally agree on is cases that are clearly anomolous should be pointed out as such, either by saying "We used the previous year's data because...." or "We used last year's data for consistancy, but....". Either is fine, but using it as an example of the overall work, I think, is improper as it really isn't representative of what the survey is trying to show, ie the current set of best tracks for players.

And Dean, yeah it grew handle, and it also lost Monmouth the pretty much the same amount of money they did the year before and came too close to shuttering the track. I'm not sure that's an example you really want to trumpet.

[DeanT]

The Elite Meet almost shut down the track? Do you have a link?

I think the subsidy went away, or the elite meet would have been back this season. At least that's what I have read (?)

[JustRalph]

Great thread. Al, good stuff. :ThmbUp:

Great info from all.....Dean and Chick, good debate. This is a good thread that makes me think. Thanks to those who work for nothing to compile the data :jump:

[Robert Goren]

"Elite" meet my foot, I have all ready bet more there this year than last. The only thing Monmouth showed last year was that if somebody else is willing or force to pay the bill for higher purses you can get some growth in handle, but nowhere near enough to pay for the return of even a part of those forced payments. Too bad they did not use the money to directly aid the handicappers. That would have produce even greater growth in handle. One has to look no further than the top rated track to see how thing should be done and they do it without taxpayer and casino money. If Hana wants to editorialize, it should be praising them. That the track model that needs to be copied.

[castaway01]

"Elite" meet my foot, I have all ready bet more there this year than last. The only thing Monmouth showed last year was that if somebody else is willing or force to pay the bill for higher purses you can get some growth in handle, but nowhere near enough to pay for the return of even a part of those forced payments. Too bad they did not use the money to directly aid the handicappers. That would have produce even greater growth in handle. One has to look no further than the top rated track to see how thing should be done and they do it without taxpayer and casino money. If Hana wants to editorialize, it should be praising them. That the track model that needs to be copied.

Just ignore Mr. Goren, he posts the same thing about Monmouth on every thread. He acts like those in charge of Monmouth, who were fighting just to keep the track open, could have just cut takeout 50 percent instead. Right, sure. He's very concerned about the 0.0001% of NJ's state budget that Monmouth's budget consisted of---we're so glad one troll in Nebraska is looking out for us in New Jersey. He lives in a fantasy world.

Thanks for the hard work HANA. While the ratings aren't perfect, they're excellent uses of the best data you had at the time. The whole chart is very useful to horseplayers.

[Robert Goren]

Just ignore Mr. Goren, he posts the same thing about Monmouth on every thread. He acts like those in charge of Monmouth, who were fighting just to keep the track open, could have just cut takeout 50 percent instead. Right, sure. He's very concerned about the 0.0001% of NJ's state budget that Monmouth's budget consisted of---we're so glad one troll in Nebraska is looking out for us in New Jersey. He lives in a fantasy world.

Thanks for the hard work HANA. While the ratings aren't perfect, they're excellent uses of the best data you had at the time. The whole chart is very useful to horseplayers. I am so glad that Taxpayers and the casinos have all that money to throw around. And what did they get for their money, a some praise from HANA and a racing industry worse off than it was before the beginning of 2010.
The HANA Chart is useful, But one things they say they are fighting for is lower takeout, the one thing that the horsemen and race tracks in NJ refuse to even consider.

[DeanT]

I actually find rating, or reading ratings of tracks (whomever does it) difficult.

For every track, or a portion of it that I like, people will hate.

I love Keeneland. Great take, they are nice to people on track. But people hate them because of poly.

I love Saratoga. The take on tris is 26% and was only raised 1% to pay for OTB's which do not exist any longer. People are not going to like them - ergo hate them- for that.

I loved the elite meet at MTH. A track actually sinking 35M into putting on a good product with full fields, and 15% take on some bets before it was the thing to do? Cool. Pennsylvania tracks are charging 31% rakes and they have not had $35M in subsidies, they have had 2 billion in subsidies.

I love Del Mar, but they chit on us with 23% rakes for exactas in seven freaking horse fields that Kreskin could not make money on at the end of the year.


Thank god Bill Weaver of HANA does the track ratings, because he has about as much bias as the least biased person in the Universe.

[Igeteven]

It makes me sick to see this, yes, I love the tracks here in California and they drove me away to play some where else, I just hope the Hana boys, Roger W and Andy A, can make changes to improve the game out here.

Right now, all 3 tracks are at rock bottom, there are people out here, that support Hana and prey they can make a huge change to bring back Ca Racing where it belongs.

Right now, it's like an old house waiting to be bulldozed, If these boy didn't care Cal Racing would be done with.

Good Luck and I hope this chart puts the fear of God into the TOC and the CHRB

[highnote]

I do think Chick and Al have valid points.

Maybe there needs to be an addendum attached to the track ratings. Or perhaps someone like Al could write a rebuttal, or whatever, to clarify things. Or perhaps that is the role Paceadvantage.com is playing right now.

Keeneland is subsidized by their horse sales -- and if I'm not mistaken, Keeneland is a non-profit. Not many other tracks are fortunate to have this kind of a setup. It's neither good nor bad. It's just the way it is and there is nothing stopping another track from following their model. It's certainly good for horseplayers who are looking for takeout value.

Monmouth may get a high rating and it can be argued whether it is a valid representation of the current meeting, but no one is forcing anyone to bet MTH because of the ratings HANA has printed.

Personally, before I bet a track I check the takeouts and am aware that the HANA ratings are only a starting point and not the final word -- albeit, a great place to start.

That said, for my money, the HANA ratings are worth every penny. :)

[rwwupl]

I think this is the best version put out so far, and "kudos" to all who contributed. I understand there will be updates and additions to follow in due time.

That being said, I wonder if ranking the tracks 1,2,3...in order is the best way to hi-lite and promote it. People will always find fault because of their home track bias and nit-picking will follow, fair or not.

It is a great handy guide, full of information, without the 1,2,3 rankings and could be promoted as the "HANA Guide" for racetracks.

Maybe the rankings and the controversy are good under the anything that brings publicity is good rule, or maybe not. The rankings will always draw criticism and charges of bias because the criteria itself is somewhat a matter of opinion and open to debate. What is good for one person is not necessarily the cup of tea for another.

Perhaps a letter grade of A, B, C, D, F would be more appropriate, serve the purpose,with less controversy.

I think it is done well as is, but I notice the "Rankings" always spark controversy, maybe that is good, or maybe not.

After the first week, the rankings are all but forgotten(except for #1) when the information content is used as a reference guide

Any thoughts?

[Robert Goren]

If they ranked tracks by a letter, they had have to grade "on scale" as they do in college otherwise they would all get a "F".

[OTM Al]

I think this is the best version put out so far, and "kudos" to all who contributed. I understand there will be updates and additions to follow in due time.

That being said, I wonder if ranking the tracks 1,2,3...in order is the best way to hi-lite and promote it. People will always find fault because of their home track bias and nit-picking will follow, fair or not.

It is a great handy guide, full of information, without the 1,2,3 rankings and could be promoted as the "HANA Guide" for racetracks.

Maybe the rankings and the controversy are good under the anything that brings publicity is good rule, or maybe not. The rankings will always draw criticism and charges of bias because the criteria itself is somewhat a matter of opinion and open to debate. What is good for one person is not necessarily the cup of tea for another.

Perhaps a letter grade of A, B, C, D, F would be more appropriate, serve the purpose,with less controversy.

I think it is done well as is, but I notice the "Rankings" always spark controversy, maybe that is good, or maybe not.

After the first week, the rankings are all but forgotten(except for #1) when the information content is used as a reference guide

Any thoughts?

It is true that any ranking is going to make controversy, but I'd much rather see numbers and methodology than letter grades. I hate letter grades. They are too subjective and seem grade-schoolish. A fixed ranking system is something the National Research Council tried to avoid last time around when they released ratings of doctoral programs. Instead they built confidence intervals around the estimates and gave a range rather than a fixed number, so the Chemistry program at school X would be rated somewhere between say 3 and 12 rather than say a fixed 8th. Of course schools either cited the top number or took an average, both things they were told not to do, so there really is no way around it. People dig lists and like to argue about them.

[chickenhead]

If they ranked tracks by a letter, they had have to grade "on scale" as they do in college otherwise they would all get a "F".

There is a lot of truth to this.

I think the original intention back when these were first talked about were more related to a report card rather than ranking. But for the perception of letter grades issue just a numeric not tied to anything makes more sense. Letter grades carry a lot of baggage with them.

If you're a track making a change that makes these things being measured more positive for players (as defined by the system), the scoring system says "good change" by giving you a higher score, making a bad change gets you a lower score. It doesn't really matter if you "pass" another track and move up the rankings or not. And it doesn't really matter whether someone thinks that number is an A or an F. Good changes are good, bad changes are bad. Numeric score goes up or down. Which is really all they are trying to do.

But yeah, lack of letter grades or having the numerics tied to any kind of scale (like out of 100 score) is probably why the rankings get so much play.
Even though, yes, if a coalition of people decided to tie this to an absolute scale of the best we can imagine, dream scenario being an A, Keeneland, the #1, might be given a D+ or something.

[highnote]

Keeneland gets the #1 rank, but perhaps the rankings need to rank the tracks relative to the ideal track.

If a 10% take and 10 horse fields is ideal then what is Keeneland's rank?

One problem Keeneland has is that if they lower their takeout other tracks who simulcast their signal complain.

Unfortunately, it will be many years before tracks approach the ideal -- if ever.

[Charli125]

Keeneland gets the #1 rank, but perhaps the rankings need to rank the tracks relative to the ideal track.

If a 10% take and 10 horse fields is ideal then what is Keeneland's rank?

One problem Keeneland has is that if they lower their takeout other tracks who simulcast their signal complain.

Unfortunately, it will be many years before tracks approach the ideal -- if ever.

Every time this is discussed we arrive at the same problems.


If we rate them compared to the ideal track the best track will have a D, maybe.
If a track lowers takeout, they risk their signal not being accepted.
Signal fees vary by who they're being sold to, and are secret. We can never know for sure what signal fees are being charged.
It's really difficult to do these ratings in a way that is helpful to both the players and as a guide for the tracks to know where they can improve because we can never know all of the facts.

[andymays]

My feeling is that the tracks should be listed in alphabetical order with a category for HANA rankings.

The rankings cause people to judge the whole thing with most being upset by the ranking of their favorite track.

It should be sold as an excellent and informative resource! Then let everyone decide for themselves.