
Google Site Search  Get RDSS  Sartin Library  RDSS FAQs  Conduct  Register  Site FAQ  Members List  Search  Today's Posts 
RDSS 2.1 Racing Decision Support System  The NEW Version! 

Thread Tools  Display Modes 
04132018, 03:53 PM  #81 
Grade 2
Join Date: Apr 2008
Posts: 75

Thanks Ted.

04252018, 09:10 PM  #82 
Grade 2
Join Date: Apr 2008
Posts: 75

Profit Line Discrepancies
Ted, I have noticed this problem, where Twins Spires profit line does not agree with the RDSS profit line. Happens about 12 times a race card and usually is just a point difference e.g. 1 becomes a 2 or vice versa. Sometimes it may be a 2 point difference. That is more problematic.

04252018, 10:01 PM  #83 
Grade 1
Join Date: Apr 2012
Location: Valley Stream NY
Posts: 1,289

Races with entries also present a problem since both parts show the same rank when in fact on the actual profit line sheet they are not. The sheet also shows what the odds should be if 1 part of the entry is a scratch.
Tim
__________________
Trust but verify 
04262018, 10:44 AM  #84 
Grade 1
Join Date: Jun 2005
Location: Huntsville, Ontario, Canada
Posts: 7,253

Please click on the 'expander' icon [+] on the Entries screen above the Scratch column at left to show several lines of info for each Entry. Under the PL column you can see the actual percentage probability figure supplied by TS for ProfitLine for each horse. If you hover the mouse over that number, you'll see an elaboration (probability and implied odds). RDSS simply ranks these probability numbers and uses those as the rank. If the rank shown on the TS ProfitLine screen is other than RDSS' rank, then they are ranking their ProfitLine differently than their own probability (and thus odds) numbers show. Which is fine! But inconsistent.
Which would you rather have: the rank shown by TS, or the rank computed by the probability/odds numbers given by TS? Alas, I do not have the PL numbers for the PRX race in question above, else I could illustrate the point. If someone could email me their PRX April 23 ProfitLine downloaded file (folder C:\Rdss2.1\Data\ProfitLine\2018\20180423\20180423PRX.htm), I will incorporate it into my readouts and repost screenshots to illustrate. Or  Pat (Old Arkie Gal), perhaps you could simply post the 'expanded' Entries screen showing that expanded ProfitLine data for the race in question and we could continue the enquiry? Also, Tim (Lt1) if you or someone can post an example of the 'expanded' Entries screen showing where each half of a Coupled Entry has a different ProfitLine number/rank  I would be obliged. In ANY event, my plan is to move AWAY from the detailed PL data available from the TS ProfitLine page shown above (since no one seems to pay attention to it, and when it diverges with TS' ranks, people complain)  to a simple listing of PL Odds and the resulting computed ranks (or tied ranks) which is now available without logging in to TS via the new tote system from TwinSpires Mobile (the former BRIS Supertote). I just thought that a ProfitLine with probability = odds of 2.91 should be ranked higher than a PL with odds of 31 though they likely are both rounded to 31 and given the same rank. (In a similar way to how there's a difference in a DRF fifthsbased time of 45.1 fifth actually being 45.39 tenths and you not knowing that it should probably be shown as 45.2 fifths but got 'rounded down' Just saying ...) Thoughts, comments? Ted
__________________
RDSS  Racing Decision Support System™ Follow on Twitter @SartinMethod 
04262018, 11:56 AM  #85 
Grade 1
Join Date: Jul 2008
Location: Carlsbad, California
Posts: 8,888

Expanded
Hi Ted
I sent you the file from Parx 4/23 on a email. Tim I am not sure about what Twinspires shows on their Profit Line sheet I never go there , I just go by what RDSS shows , so I can't comment on any difference if both or only one half of a entry run Here is race 2 from 4 24 Expanded 
04262018, 12:08 PM  #86 
Grade 1
Join Date: Jun 2005
Location: Huntsville, Ontario, Canada
Posts: 7,253

Bill, could you show the Race 5 we are discussing above? Thanks.
EDIT: Never mind Bill  I got your file and will post the Race 5 expanded screen myself. Ted
__________________
RDSS  Racing Decision Support System™ Follow on Twitter @SartinMethod Last edited by Ted Craven; 04262018 at 12:11 PM. 
04262018, 12:09 PM  #87 
Grade 1
Join Date: Jul 2008
Location: Carlsbad, California
Posts: 8,888

Entry
Hi Tim
Here is a race from Parx on Tuesday The 1A was scratched Here are the RDSS screens with the 1A scratched and with the IA back in Either way RDSS gives the entry an 18.52 % PERCENT AND A 4.4/1 ODDS But like I posted above I never look at the twinspires PL "sheets" so I can't comment on what they show 
04262018, 12:23 PM  #88 
Grade 1
Join Date: Jun 2005
Location: Huntsville, Ontario, Canada
Posts: 7,253

Here is the expanded screen for the race in question. Horse #3 has a slightly higher PL probability/odds value of 15.51 (rounded to 15.5) or odds of 5.41 compared to Horse #1 which has probability of 15.43 or odds of 5.51. That is why RDSS ranked the #3 as rank 2 and the #1 as rank 3, which I agree is different by 1 rank position from what is showing on the TS ProfitLine screen where BOTH are shown with ProfitLine odds of 61 (i.e. rounded UP) and the ranks are reversed for some reason. All these numbers come from the same TS ProfitLine screen (just under the surface). That's the technical answer for where RDSS gets its ranks.
Which numbers would you rather have? Here is the TwinSpires ProfitLine page shown above with slightly different ranks. Ted
__________________
RDSS  Racing Decision Support System™ Follow on Twitter @SartinMethod 
04262018, 12:32 PM  #89  
Grade 1
Join Date: Jun 2005
Location: Huntsville, Ontario, Canada
Posts: 7,253

Quote:
Tim  if so, can you elaborate, maybe display, the different online ProfitLine source. The source I use (the TS ProfitLine display aboe, always gives identical PL values for both horses in a coupled entry. Again, as I mentioned above, I intend with the next update to simply fetch the PL Odds (in my above example) from the TS Mobile odds stream, and abandon the finegrained ProfitLine probability values I have been using. Thus, in the example above for PRX Race 5 April 23, we'd see BOTH horse #3 and #1 show as 61 (and NOT 5.51 and 5.41) and BOTH ranked identically. I will retain the existing means of fetching ProfitLine by logging in to TS, but the new automated procedure will happen without any initial login required. (FWIW  I will also be adding the BRIS Prime Power number as a new external 'power number'  ranks, gaps and the number itself, and integrated into Rx if one chooses). Ted
__________________
RDSS  Racing Decision Support System™ Follow on Twitter @SartinMethod 

04262018, 12:44 PM  #90 
Grade 1
Join Date: Apr 2012
Location: Valley Stream NY
Posts: 1,289

Hi guys. I'm not good posting screen shots but here's an example of what I'm talking about. In tomorrows 6th race at Bel we have the 1 entry. Although neither is ranked in the top 5 they are ranked on the sheet. In this case the 1 is ranked #6 while the 1a is 12th. Ml for both is 61 however the PL odds are 181 for both if they both run. If the 1a were to scr the 1 remains 181 however if the 1 scr the 1a would be 501. This is the way all entries are reflected on the actual sheet regardless of their rank. So when 1 of them scr make sure to check if the ranks change in the grid.. Even the Bris pps power #s ranked them differently. For example in this race the 1 is ranked 10th while the 1a is ranked 6th. Ted I don't know if anything can be done about this so it looks like it's something we have to live with. For those who watch the races on the TS screen sometimes when 1 of the entry is top 3 rated you will see it in the Handicapping box. I don't know if this helps or not.
Tim
__________________
Trust but verify 
Thread Tools  
Display Modes  

