Assessing horse performance

[mardigras]

I assess performance by assessing the time a horse runs against what is expected to see what level of performance a horse has run to. I use my own rating system.

I've read some suggest that it would be too difficult to do that if the horses are randomly running around on different tracks. Hello, that is exactly what happens. Maybe they need to brush up on their ability to analyse data.

Once you have assessed the state of the track, you can compare a horse's performance against all other performances on that track under the same assessment of track, at that distance. With the ability to assess collectively all those performances against the assessed ability of all those horses, you can determine where each performance sits relative to the rest. To the point where you can assess the relative performance of every horse to help define an ability of every horse in the country relative to the rest. And obviously that assessment can be altered by new information/performances.

For me, if I couldn't determine the level of performance from one track to the next, I'd be stuffed when it came to working out who I thought was superior between two horses that had been racing on different tracks.

And as for running 3 wide, I don't do subjective. Especially when I'd expect 3 wide to be just as much chance of being a positive as a negative.

[Hesi]

43 minutes ago, mardigras said:

I assess performance by assessing the time a horse runs against what is expected to see what level of performance a horse has run to. I use my own rating system.

I've read some suggest that it would be too difficult to do that if the horses are randomly running around on different tracks. Hello, that is exactly what happens. Maybe they need to brush up on their ability to analyse data.

Once you have assessed the state of the track, you can compare a horse's performance against all other performances on that track under the same assessment of track, at that distance. With the ability to assess collectively all those performances against the assessed ability of all those horses, you can determine where each performance sits relative to the rest. To the point where you can assess the relative performance of every horse to help define an ability of every horse in the country relative to the rest. And obviously that assessment can be altered by new information/performances.

For me, if I couldn't determine the level of performance from one track to the next, I'd be stuffed when it came to working out who I thought was superior between two horses that had been racing on different tracks.

And as for running 3 wide, I don't do subjective. Especially when I'd expect 3 wide to be just as much chance of being a positive as a negative.

3 points

1.  How do you quantify, what is expected

2.  Even though horses are expected to be run 100% fit, we know that does not happen.  Trainers even acknowledge a horse is only 90% etc etc.  How do you factor that in assessing actual vs expected

3.  Why do all the commentators, in most races, comment negatively on a horse being 3 wide without cover, but more positively on a horse being 3 wide but with cover.  You hear it most races

[mardigras]

4 minutes ago, Hesi said:

3 points

1.  How do you quantify, what is expected

2.  Even though horses are expected to be run 100% fit, we know that does not happen.  Trainers even acknowledge a horse is only 90% etc etc.  How do you factor that in assessing actual vs expected

3.  Why do all the commentators, in most races, comment negatively on a horse being 3 wide without cover, but more positively on a horse being 3 wide but with cover.  You hear it most races

1 & 2

There is two parts to what I do. Part 1 is to assess the performances after the races. Part 2 is to assess (predict) the performance of the horse(s) before the races.

Part 2 includes an aspect of confidence around things such as - condition of track, distance suitability, fitness based on trends shown - to assess where the horse might be for this run relative to my assessment of its ability.

Part 1 ignores my assessment done in Part 2 - and is solely an assessment of the horses performance on the day. Since those assessments are what are used to perform Part 2.

Say if you assess the state of a track as a 4 (ignore penetrometer etc). And you rate horses between 0 and 10.

Then for a 0 rated horse over 1200m, you can define what the time expected would be, right through to what the time would be for a 10 rated horse.

And on one track, those times might be a lot different to what they would be on another track.

When the horses run, you can assess where in that spread each horse ran, effectively giving it a rating for that run.

I don't factor in  horse not being fully fit since at that point I'm only assessing the run. The run isn't the overall assessed level of the horse, as I look at the lifetime of assessments to determine my overall ability assessment. And number of starts feeds into my confidence aspect of that assessment as to whether I expect it to perform to the assessment I have. So if I have low confidence that the horse is going to run up to my assessed ability, then that affects assessed price.

I don't want to factor in my expected (pre race) into my run assessment (post race), since the actual run is used as the basis for determining things such as fitness by way of trend analysis. So I stick with each assessment being a raw assessment of the performance.

3. I'd say because it's ingrained into the way punters and the like think in NZ. To me, without cover means the horse is less restricted (unless the rider is doing so). With cover means the horse is subjected to whatever pace changes etc occur in relation to the horse in front of you. As is the case for any horse following or surrounded by horses. If I saw a real world study that suggested things were different, I'd possibly look at it - but ultimately I doubt I'd change anything since it would require some subjective factor as to what that did or didn't cause (from a post race perspective), and again, I wouldn't know how to consider it from a pre-race perspective.

And for me, the best real world analysis would come down to analysing expected performance versus actual performance to determine whether there was a correlation between a horse running 3 wide without cover having a detrimental impact on performance (actual v expected) against say those horses running 3 wide with cover (actual v expected). I can't do that analysis because I don't have the information on the position of each horse in the race. And even then - it would become a population statistic which would still raise questions as to how to apply that to an individual horse pre race.

[Hesi]

For the benefit of our know all friend(who never did answer how many identities he has, Noodlum included), who thinks what you do is boring, are you able to 

1.  Give an indication of time spent on maintaining the system each week.  I realise a lot of initial work went into the set up.

2.  The average return on investment

[mardigras]

20 minutes ago, Hesi said:

For the benefit of our know all friend(who never did answer how many identities he has, Noodlum included), who thinks what you do is boring, are you able to 

1.  Give an indication of time spent on maintaining the system each week.  I realise a lot of initial work went into the set up.

2.  The average return on investment

1. No time spent maintaining it day to day as such (except see below). Operational time (ensuring data is kept up to date and config set up day to day), half an hour a day. A little more catching up if I go away on data.

2. It's a little more difficult to define. Straight win betting in the 25 - 35% ROI, lately closer to 100% but it fluctuates. But I do a fair bit at times of laying full markets at which point it becomes a little cloudy as to what is the investment - eg. If I lay a market to an exposure of say $1,000, and then increase that to say $2,000 after already being in a non-risk position, is that investment of $1,000 or investment of $2,000? I've done markets with exposure well into the 100s of K, but the risk hasn't been 100s of K.

And also, with laying full markets, your ROI will always be much lower as you are dealing with markets usually in a highly competitive state. On some rare races I will achieve a 20% ROI, on the majority a 0-3% ROI, and on some I will lose if my assessment of where prices will go is beyond an acceptable limit.

On the topic of maintenance - on 1, I do operate a continuous improvement cycle. So every 2 - 3 years, I will review algorithms, the assessment of performance processes etc, to determine whether things are operating well or whether I have missed something. New tracks can take some effort to ensure assessments on tracks are being done as well as possible.

When I do that, that could be a couple of hundred hours revisiting things running what-if analysis and various iterations. That period is usually something I look forward to as it gives me a chance to ensure my brain is being challenged which is a good thing.

Not a lot of time to get bored day to day on it. Since I don't spend much time on it. I don't really do much around racing these days. Rarely go, rarely watch. usually only watch if I see something on here and decide to check out what some have said for interest sake. 

 

Edited December 14, 2020 by mardigras

[pete]

35 minutes ago, mardigras said:

1. No time spent maintaining it day to day as such (except see below). Operational time (ensuring data is kept up to date and config set up day to day), half an hour a day. A little more catching up if I go away on data.

2. It's a little more difficult to define. Straight win betting in the 25 - 35% ROI, lately closer to 100% but it fluctuates. But I do a fair bit at times of laying full markets at which point it becomes a little cloudy as to what is the investment - eg. If I lay a market to an exposure of say $1,000, and then increase that to say $2,000 after already being in a non-risk position, is that investment of $1,000 or investment of $2,000? I've done markets with exposure well into the 100s of K, but the risk hasn't been 100s of K.

And also, with laying full markets, your ROI will always be much lower as you are dealing with markets usually in a highly competitive state. On some rare races I will achieve a 20% ROI, on the majority a 0-3% ROI, and on some I will lose if my assessment of where prices will go is beyond an acceptable limit.

On the topic of maintenance - on 1, I do operate a continuous improvement cycle. So every 2 - 3 years, I will review algorithms, the assessment of performance processes etc, to determine whether things are operating well or whether I have missed something. New tracks can take some effort to ensure assessments on tracks are being done as well as possible.

When I do that, that could be a couple of hundred hours revisiting things running what-if analysis and various iterations. That period is usually something I look forward to as it gives me a chance to ensure my brain is being challenged which is a good thing.

Not a lot of time to get bored day to day on it. Since I don't spend much time on it. I don't really do much around racing these days. Rarely go, rarely watch. usually only watch if I see something on here and decide to check out what some have said for interest sake. 

 

All this from someone who is only a SQL DBA 🙄

[mardigras]

8 minutes ago, pete said:

All this from someone who is only a SQL DBA 🙄

It seems the chief feels the need to answer for others as well. Answering for curious as well. I guess when there has been 82k posts from the chief and wandering eyes, the desire is strong , to answer to every thread instead of waiting for those asked to respond.

[pete]

He accused me of being a SQL DBA as well. Hilarious. 

[mardigras]

Just now, pete said:

He accused me of being a SQL DBA as well. Hilarious. 

As you said - maybe that's the only other occupation he knows. Nothing wrong with anyone being a DBA, but it certainly isn't represented anywhere on my CV. I'm almost feeling sorry for him - as I do with many that are somewhat challenged.

[Hesi]

SQL DBA

Structured Query Language Data Base Administrator?

Well most of us probably thought it was a term out of Gladiator

 

[Hesi]

10 minutes ago, mardigras said:

It seems the chief feels the need to answer for others as well. Answering for curious as well. I guess when there has been 82k posts from the chief and wandering eyes, the desire is strong , to answer to every thread instead of waiting for those asked to respond.

Considering he is any one of upwards of a dozen different posters, that could all get a bit confusing.

Don't laugh if he is prepared to be Wandering Eyes, Boxie and Noodlum, then why would you stop there.....all in the name of manufacturing debate.  And the guy has the nerve to accuse you Pete of setting up several other accounts on his site

All part of the technique, best form of defense is attack

[Hesi]

So, sounds like you earn quite a decent living from wagering Mardi.

I presume it is a constant battle to stay from being restricted on the different sites

[mardigras]

3 minutes ago, Hesi said:

SQL DBA

Structured Query Language Data Base Administrator?

Well most of us probably thought it was a term out of Gladiator

 

In the Microsoft world of SQL Server, that would be largely around

installation/clustering/config

schema definition/indexing

security

maintenance/backup/recovery/disaster recovery/log shipping

migration/deployment

Not one of those things have I ever done in my working career - for anyone. On that basis, I'd certainly agree with anyone that was to suggest that I'd likely make a pretty shitty DBA. 

 

[mardigras]

12 minutes ago, Hesi said:

So, sounds like you earn quite a decent living from wagering Mardi.

I presume it is a constant battle to stay from being restricted on the different sites

I wouldn't call it a decent living - but it certainly gives me enough to do my travels etc. I'm fortunate enough that I don't need income from that so I do it for the challenge of being able to be successful. And a few close mates get information from me most days.

I stopped working about 17 years ago to mainly generate money from punting. But my brain needed some stimulation once the development was done. So 3 years ago I returned to do some work (part time) in BI. I don't do that for the money either, but keeps my brain active. 

You don't get restricted on betfair. And that is one of the few sites where they have decent APIs for information and bet placement. All that happens with them is the more you tend to profit and the faster you profit, you pay them more fees. Which I do. Since they are peer to peer, they aren't generally the other side of the transaction, more a clip the ticket process (a bit like tote). if they think you aren't paying them enough under their rules, they can increase the rate they clip the ticket for you.

Edited December 15, 2020 by mardigras

[pete]

29 minutes ago, mardigras said:

In the Microsoft world of SQL Server, that would be largely around

installation/clustering/config

schema definition/indexing

security

maintenance/backup/recovery/disaster recovery/log shipping

migration/deployment

Not one of those things have I ever done in my working career - for anyone. On that basis, I'd certainly agree with anyone that was to suggest that I'd likely make a pretty shitty DBA. 

 

Same with me. I'm a developer and will be until I decide I don't want to do it anymore.

[mardigras]

53 minutes ago, pete said:

Same with me. I'm a developer and will be until I decide I don't want to do it anymore.

I've been on the development side of the fence my whole life.

I found this pretty funny from Noodlum - or was that The Chief? Is he talking about his own posts?

Well people are quick to try and bring the successful down by casting unfounded aspersions

[pete]

1 minute ago, mardigras said:

I've been on the development side of the fence my whole life.

I found this pretty funny from Noodlum - or was that The Chief? Is he talking about his own posts?

Well people are quick to try and bring the successful down by casting unfounded aspersions

Saw that. He seems to now find calling you an ex DBA amusing.

Pathetic.

[Hesi]

1 hour ago, mardigras said:

I've been on the development side of the fence my whole life.

I found this pretty funny from Noodlum - or was that The Chief? Is he talking about his own posts?

Well people are quick to try and bring the successful down by casting unfounded aspersions

You cannot be serious, he has spent the last 6 months trying to discredit the successful handling of Covid in this country, including a section devoted to just that, despite the plethora of expertise available to them