PDA

View Full Version : spot plays by track


JACKSON77
01-13-2008, 10:05 PM
Wondering how effective it is to break down spotplays by track. I hate to do all that work just to find the spot plays are good one year and suck the next. Do you fellas see any correllation to speed plays at the same tracks year in and out. Do your spot plays suck at the same track consistantly. YOu know..ones like top speed horse where there is little speed and it has some fr3. My play returns a profit over all tracks. I wonder if it would be consistantly better if I chuck tracks that it did bad at this year. Do you guys understand what im talking about here. MY data base is small .. like 20k races total. One would need about 5 years data on each track to know for sure. Seems like a helpful thread, though im sure it has been thought of before. Thanks

ryesteve
01-13-2008, 10:12 PM
I don't know if 5 years worth of data would help, because the nature of the surface can change (especially with more tracks installing poly) which would probably affect your results. Depending on what your winning pct is, you can get a feel for how you're doing from track to track with far fewer races.

JACKSON77
01-13-2008, 10:53 PM
My spot has made a 1.10 roi for 2 solid years now and has 1 play every 2 days or so. Plus i get 8% on avg rebate. Its over all tracks and I found lot of others that did very well and have lots of plays. I just dont trust them that well. Some made a 1.15 for the year but had months of .7... which would devastate a bankroll. I pretty sure with my rebate im unstoppable but this is real money. I also thought of keying my longshot plays in pick 4s for a shot at increasing my roi since they always seem to pay twice what they should. Have you tried that. I have one that kicks out $20 horses, which usually mean big payoffs as opposed to correct odds pay offs in pick 4s. But would require a huge bankroll if I had loss streaks in the keying to the pick 4 scenario.

JACKSON77
01-14-2008, 07:34 AM
Anyone with a huge base ever try to break yours down for consistancy by track?

DanG
01-14-2008, 07:44 AM
Wondering how effective it is to break down spotplays by track.
My experience is track / surface specific is the way to go.

As a friend is fond of saying…”There are no two tracks exactly the same…the surface, maintenance right down to the hot dogs and beer all have a unique element.”

I use a wild method with spot plays that I probably wouldn’t recommend to people looking to avoid a straight jacket. :eek: I currently have around 85 separate plays in rotation and pull them in and out of circulation by performance. The unusual twist I pull them out after a strong run and bet them strong after underperforming.

The same contrarian logic that some on Wall Street have employed for many years. My records have shown reasonably predictable ranges of peaks and decline when I measure the average performance / adjusted for standard deviation.

Long story short…Yes, track specific I’ve found to be much more manageable and makes record keeping extremely accurate.

Best of luck!!!

shanta
01-14-2008, 12:11 PM
The unusual twist I pull them out after a strong run and bet them strong after underperforming.


Light bulb moment here Dan. Thank you!

JACKSON77
01-14-2008, 05:16 PM
I have seen something like that on hsh and it makes sense. As soon as someting starts to pay well thenit getsjumped on by every software package on here. thus the roi decreases. The only way to avoid that is get really obscure or go contrarian. I t makes perfect sense to me because you know that over time, meaning years, that the play is strong. You should crap yourself when it goes to 90 cents for 2 months. You know the next 2 months should be 110. I bet your notes show the hit rate staying the same and the payoff dropping. I just want to eliminate some tracks where my speed bias horses dont do well.. or tracks with weird layouts.. charlestown..ect.. but only if it sucks yearly. Im taking it that you say dump the tracks where it hasnt done well this year. what are signs that its the track and not just a swing. Probably across the board failure and not just the win end.

russowen77
01-14-2008, 05:47 PM
Almost all life is cyclic. I had spaced that our in regard to handicapping. Thanks so much DanG.

JACKSON77
01-14-2008, 06:18 PM
I will work 3 of my best back 6 months and see what they did... With my 10 percent win rebate I think I still will be ahead.

trying2win
01-14-2008, 06:53 PM
[QUOTE=JACKSON77] My play returns a profit over all tracks./QUOTE]

Yeah right! I get a kick out of new PA members with a few posts claiming to be big winners, then they ask for advice. My estimate of the 'real winners' out of all the posters at this site BRAG ADVANTAGE, er I mean PACE ADVANTAGE over the years, is in the 10 to 15 % range. And that estimate is probably generous, because of a fair number of sharp handicappers who I believe to be long-term winners. I pay attention to their posts here. Helps to trust one's intuition to separate the 'real winners' from the 'legends in their own minds' posters here.

The 'real winners' in my opinion here at this site generally don't brag. They demonstrate their ability as a handicapper with their insightful, brilliant knowledge and wisdom. I admire and thank them for their willingness to share their handicapping expertise with others.

JACKSON77
01-14-2008, 08:10 PM
whats his problem? Please everyone add thoughts to this post. Lets learn from those that are positive thinkers and not the little negative ones.

jma
01-14-2008, 08:36 PM
My experience is track / surface specific is the way to go.

As a friend is fond of saying…”There are no two tracks exactly the same…the surface, maintenance right down to the hot dogs and beer all have a unique element.”

I use a wild method with spot plays that I probably wouldn’t recommend to people looking to avoid a straight jacket. :eek: I currently have around 85 separate plays in rotation and pull them in and out of circulation by performance. The unusual twist I pull them out after a strong run and bet them strong after underperforming.

The same contrarian logic that some on Wall Street have employed for many years. My records have shown reasonably predictable ranges of peaks and decline when I measure the average performance / adjusted for standard deviation.

Long story short…Yes, track specific I’ve found to be much more manageable and makes record keeping extremely accurate.

Best of luck!!!

Good point there Dan! Your logic makes sense, but it does take a knack for knowing when to jump off the bandwagon and when to jump back on. I find the same thing with my trainer plays. The minute everyone jumps on, it's either time to stop playing or to find a new way to make it profitable. Not easy, but possible! You seem to have it down better than I do though.

njcurveball
01-14-2008, 09:10 PM
whats his problem?

You learn that using the Ignore feature here improves your experience. :ThmbUp:

JACKSON77
01-14-2008, 09:50 PM
I have removed unprofitable surfaces. I have found to only use factors that scratches wont impact. Such as the best figure in a catagory.. Although I have found having the 4th and 5th best horse is amazingly profitable if the odds are right.

Ihave seen huge odds swings on the tracks that the ptc has. I suspect its due to lots of wagers waiting till 30 seconds to slide into the pools. Thats scary if everyone is using it. You could be requiring 5 to one and suddenly your bets in and last flash its gone for 5 to one to 5-2. That happened to me twice last week. Too much drop for normal wagering. I started to think they were betting after the bell.

bigmack
01-14-2008, 10:29 PM
I use a wild method with spot plays that I probably wouldn’t recommend to people looking to avoid a straight jacket. :eek: I currently have around 85 separate plays in rotation and pull them in and out of circulation by performance. The unusual twist I pull them out after a strong run and bet them strong after underperforming.
Make no mistake about the straight jacket that bassman DanG alludes. This cat puts a monstrous amount of work into what he does and reaps the rewards for his efforts. Most lack the fortitude to work on that level.

If you're a player with a reasonable amount of accumen you know that spot plays work. To even the most ardent opponent of computer programs I ask; how can you not see the alchemy to quickly identify conditions to your specifications and have them presented to you for every horse they fit for every race, at every track, for every day of racing, for $4/day?

DanG
01-15-2008, 07:01 AM
Does take a knack for knowing when to jump off the bandwagon and when to jump back on. I find the same thing with my trainer plays. The minute everyone jumps on, it's either time to stop playing or to find a new way to make it profitable. Not easy, but possible!
It’s far from a perfect science, but it’s improved my net by 15-20% on average. When I mentioned the straight-jacket that Big-Mack was alluding to I wasn’t kidding. To take a play out of circulation after 30%+ returns and replace it with a 30% loser I admit takes a bit of madness. The ‘straight-jacket comes into play when both plays have II tops and / or lows. :bang:


I take ZERO credit for the idea as it was my brother who showed me the wisdom in it. He has the brain power of any three people I know, but does not like computers. He worked out a long hand method of charting trainer / rider performance and noticed how there were cyclical peaks and valleys that were “somewhat” :rolleyes: predictable. The same basic theory that chartists have applied to Wall Street for decades. (He could not make the charting of barns and riders show profit btw for what its worth)

Spot plays that are coming from unusual angles take long hours to develop as those who have tried can attest. If you’re loading in the best last fig or whatever other “mainstream” factor you fancy you may as well try another approach right now. You also must factor roughly a 10% average decline for most plays as some guy in Istanbul is crunching similar data and smoking a 9 foot pipe while he’s doing it. :cool:


I believe it was Jackson77 who asked how one knows to dump a track.

Well…Once a 12 month profile is established (I use 30 day cycles) then “break out” points can be assign; again similar to what chartist do. If a play has a median ROI of $1.15 and a standard deviation of .20 cents I multiple the deviation by 1.4 to establish thresholds of when I will pull and / or place it in rotation. (Give or take...$0.87 & $1.43)


BTW: This isn’t exactly how I track plays, but makes for an easier example this time of the morning! (I use medians, nets, standard deviations and keep a close eye on place %’s to eliminate plays that are truly just getting “lucky”. At the same token putting plays in that just so happen to be losing some dirty pictures.)

The following rough guidelines also apply off the top of head…




Very conscious of track profile changes, especially on plays that feature position / velocity.
I rarely pull a play in mid-month once the decision is made to put it in and vice a versa regardless how it starts the cycle.
I’m VERY aware of factor overlap in play creation.
The best plays I have would not pass the “logic” test of most horseplayers. (I can’t answer as to why, but I guarantee several plays would have many saying that can’t work.)
Seasonal tracks are treating as one continuous meet. I.e. Saratoga is still charted on monthly basis for bookkeeping sake; even though each cycle is treated as unique.
Before a play is created the circuit (players who move the pools) must be evaluated. Such as contrarian angles in NY when sheet players will NOT be on the animal.
Spot plays must be segregated by hit rate and grouped with diversity such as a good mutual fund. A good mix of 9% bomb seekers some mid-priced and “blue chip” grinders has given the overall portfolio a better chance of succeeding.
It’s too early…Maybe more will come to me after some more java!!! That should be enough confusing signals for some reply’s to tell me how nuts I am!!! :D

JACKSON77
01-15-2008, 09:23 AM
Thats a great post, I have crunched my numbers on my 3 plays. I can post them going back in time to show how they have been mostly ahead for the last 6 months. One exception is one that has actually gotten killed lately that has been a huge winner for 2 years. Its hit rate dropped to 17% from an usual mid 20 %. This would maybe close to Dans jump in point.
I know this will be confusing but its 30 day segments going backwards to July. IN example the 139 plays would be all plays possible in last 6 months
spot 1
Plays wins roi win place show
139 48 1.34 1.01 1.04
110 39 1.28 110 1.06
85 28 1.03 95 98
47 20 1.28 1.03 1.00
24 11 1.42 .95 1.03

spot 2 roi
170 45 1.23 1.00 96
135 34 1.13 1.04 97
114 26 1.02 .98 .95
74 17 1.06 1.02 97
53 11 .98 1.00 92
35 6 .87 1.11 92


spot 3
plays wins roi
680 175 112 97 92
523 136 107 98 92
425 114 112 100 83
288 74 105 92 89
128 38 124 105 92

All my spots win consistantly except number 2 which held up well in place but lost someground on win end recently. I see this as strength because ith has returned 1.10 for the last 2 years. We all should be doing what Dan is doing to our spots that tank for 2 months, especially since the guys in istanbul hammer them when they are making 1.10 down to .87. Once they refigure thier numbers and jump off it will make the play go to 1.20 for a few months I bet.

jma
01-16-2008, 02:44 PM
Really interesting stuff Dan, thanks for taking the time to explain your methods.