Scoring a Par Time Stage on Practiscore

[-=VILLAMOR=-]

We recently obtained Nook tablets with the Practiscore app from one of our match sponsor. We've ran it through some mock matches before using it in two local matches. We have not had any issues and we absolutely love how much work it saves us. The only problem is that I can't figure out a way to score a par time stage using the Time Plus method in Practiscore.

The stage gives the shooter 25 seconds to knock down as many targets as he can. The shooter that knocks down the most will get the 100% value of the stage and everyone below will get a percentage based from the winner.

Has any one have a recommendation in scoring such a stage in the Time Plus method of scoring? Your assistance would be greatly appreciated.

[Bryan 45]

I haven't put a lot of thought into this, but what if all shooters are scored 100 seconds, and each target is a -2 second bonus?


You just have to make sure it's not possible for anyone to score 0 or negative. With the target arrays, would it be possible for an open shooter to have 50 hits?

[euxx]

Bryan's suggestion sounds like a good idea. Just need to ballance the bonuses with the number of targets, so it will total to 0, which would give the max points for a timeplus/points match.

Though I'd leave time at 25 seconds and unfortunately you'll have to enter sane time for every shooter. Maybe something we could improve in the future, so email your ideas to support@practiscore.com or talk to me at RC next week

[StealthyBlagga]

Bryan's suggestion sounds like a good idea. Just need to ballance the bonuses with the number of targets, so it will total to 0, which would give the max points for a timeplus/points match.

Just a small point of order. The points calculation in the Time-Plus scoring system is as follows:

STAGE_POINTS = ( FASTEST_TIME / PARTICIPANT_TIME ) x 100

Thus, all competitors MUST end up with a time-plus-penalties stage time >0 or the equation becomes mathematically insoluble.

Furthermore, if the number of targets is not impossibly huge, it is feasible that one or more competitor might clear the stage inside the par time, in which case you will need the actual clock time to differentiate between those shooters. You might therefore want to use the actual time (or par time if maxed out), and pro-rate the bonus with adequate cushion to ensure nobody gets a stage time <0.

Edited May 23, 2014 by StealthyBlagga

[euxx]

We have some magic to handle this. I've just tested it for both timeplus and timeplus/points matches and Android app calculates results correctly when final time is 0.

BTW, they used to run steel challenge matches with bonus plates and some people managed to get a negative time (which was carried over to the overall time), so it is a valid case.

[CJW]

Just to be pedantic, what is being described is not a par time stage, but a fixed time stage. A fixed time stage is essentially tied to a points based system and is a little hard to adjust for time based scoring.

I am thinking StealthyBlagga has the right idea. Treat it as a 25sec par time, score the actual time (up to 25 sec) and then add a penalty (2sec, 5sec, whatever) for every unhit target.

Or, flip side, score everybody 125 seconds less their unused time at last shot, call each hit a 5 sec bonus target (for up to 20 targets). So, a guy who clears all 20 targets in 18 seconds will get a time-plus of:

125 - (25-18) - (20*5) = 118 - (20*5) = 118 - 100 = 18 seconds

(For ease of calculation, that works out to 100 + actual time, less bonuses.)

[MarkCO]

A Par time in Time plus can't result in a score of zero. Every competitor gets the Par time entered, then add penalties, which, in Time-PLUS are added, not subtracted.

So for instance, 3 strings, 5 seconds Par time, let's assume 5 targets that require a total of 5 hits ...

Everyone gets 15 seconds entered, and then the targets are scored on a per shot miss basis. Stage winner has no misses and gets 15 seconds. Shooters A has 2 misses getting a total time of 25 (15+10). Shooter B has 10 misses and gets a stage time of 65 (15+50). Stage winner is 100%, Shooter A is 60% and Shooter B is 23%.

There is NO subtraction and no-one will zero the stage in Time-Plus. Pretty simple. I have run them in Time-Plus this way as a Standards stage.

Edited May 23, 2014 by MarkCO

[Jakobi]

^ this.

Everyone starts with the par time (fixed time) and is assesed any penaltiles for targets not hit. Points are calculated as normal.

-Jake

[euxx]

A Par time in Time plus can't result in a score of zero. Every competitor gets the Par time entered, then add penalties, which, in Time-PLUS are added, not subtracted.

Sure it is, if you use bonuses instead of penalties as Bryan suggested.

Logically it makes more sense to reward shooter for hits, rather than penalize him for targets he didn't had enough time to shoot... But I suppose it is up to the personal MD preferences, because the smaller time wins either way. Though they will affect the overall results differently in a timeplus match.

[Jakobi]

I think it's irrelevant whether the shooter thinks they're being rewarded or penalized. The question is what is the easiest way to score it? Time plus is based on the lowest aggregate time for a stage (the inverse of Hit Factor scoring where the highest score wins). You can try to do it with a bonus and subtract from the par time or you can simply count up how many targets are left standing and make that the shooters score (plus par time). Guy(s) with par times only win, everyone else gets a percentage of that. If it's a bonus you have to make sure you can't get a negative or zero time or the math doesn't work. For me, I would count the remaining targets and add that to the par time. Seems easier (from a scorekeeping standpoint) than trying to work out a bonus system.

-Jake

[euxx]

I'd suggest to try it in Practiscore and see if math work or not...

[CJW]

The difference between what is being described here and what I think the OP was actually asking for (a fixed time stage) is how the scores scale.

Take Mark's example above, and let me change the numbers a little. 25 second fix time and 5 sec per target. Clean gets you a score of 25 sec and 100%. Five misses gets you 50 seconds and a stage score of 50%...regardless of how many targets are on the stage. If there were twenty targets, then five misses (hitting 75% of the targets) only gets you a 50% score. If there are 40 targets, your five misses means you hit 35/40 = 88% of the targets...and you still get a 50% stage score.

The purpose of a fixed time stage is the linear point scale. Scoring it as time plus with penalties makes the score distribution very bottom heavy.

Now, if all you want is a stage where good shooters will be challenged to hit all the targets, then just set a low (20--30 second) par time and score it like a regular stage: actual time plus miss penalties. I would at least say you can skip any failure-to-engage penalties, though.

[MarkCO]

A Par time in Time plus can't result in a score of zero. Every competitor gets the Par time entered, then add penalties, which, in Time-PLUS are added, not subtracted.

Sure it is, if you use bonuses instead of penalties as Bryan suggested.

Logically it makes more sense to reward shooter for hits, rather than penalize him for targets he didn't had enough time to shoot... But I suppose it is up to the personal MD preferences, because the smaller time wins either way. Though they will affect the overall results differently in a timeplus match.

It actually depends on the ruleset. That would not be an option under the traditional definition of Time-Plus how USPSA and IDPA use it.

[J.Schmitt]

Jomar and I tired a variety of calculations from 1 sec to 5 seconds bonus to penalties. We are tring to make so that if a troubled shooter shoots only 50 percent of the targets they get a score that matches at 50 percent. So last year we had a shooter who hit 25 and got 100%. Another shooter got 12 and got 48%.

So this was our exact concern that we did not want to have happen. We did not want to over penalize.

Take Mark's example above, and let me change the numbers a little. 25 second fix time and 5 sec per target. Clean gets you a score of 25 sec and 100%. Five misses gets you 50 seconds and a stage score of 50%...regardless of how many targets are on the stage

Jay

[sschultz]

Jay are we going to use the tablets on thr Nordic match.

[J.Schmitt]

Yes, That is why we are tying to solve this. Jomar's spreadsheet does this stage perfectrly. Trying to find the best solution for the new system.

[-=VILLAMOR=-]

Thanks for all the imput so far. I forgot to mention that the point of the stage is to have more targets available than a shooter can feasably engage within the fixed time. Its a speed stage where different strategies of when, where, and what to engage in conjuction with each individual skills will make the difference. We don't want anyone to flat out clean it under the par time. The shooter with the most hits get 100% of the point value of the stage. If that winner hit say 30 targets and another shooter hits only 15 targets, then the shooter shot 50% of the winner and gets 50% of the value of the stage.

I played around with both the same time plus bonus for every target hit and the same time plus penalties for every target not hit. But what I immediately notice is that if I have the winner hitting 30 targets and the other shooter hitting 15 targets, it will not calculate the score of the other shooter as 50%. The system is programmed to calculate time value and not per target point value.

I love the Practiscore system. It makes scoring so much simpler and faster. It saves me from hours of manually entering scores. It would be great if we could use the Practiscore for our big match and still have our "fixed" par time stage. But with me being the score keeper, I am very much partial to Practiscore. (hint hint Jay)

[euxx]

Have you looked at "timeplus w/points" match type? There you can assign each stage a point value (the best time will get the 100% of the points for that stage) and match overall results is a total of per-stage points.

[-=VILLAMOR=-]

Yes. The point value of the stage is not the problem, it is the distribution of the points. No matter the point value of the stage, the shooter with the most hits should get 100% and the shooter with half of the winner's total hits should get 50%. But that is counting per target hit value and not time value as Practiscore is designed to do.

[CJW]

The short version is that you can't do what you want to do with TIme Plus. The basic reason is that the process of dividing times (score = best time / this time) produces a curve that will not give you the linear relationship you want (e.g., a quarter as many targets hit gives you 4 times the best time) across the whole range of possible scores, and this is made worse by not knowing what the bestscore is going to be. Also, a Time Plus result will never produce a 0% score and will instead tend to get really flat around 10–15% or so (give or take the specific numbers).

Your best option is to score this stage manually, which obviously makes match scoring harder. You could take the manual scores, take them to the stat shack, put them into a spreadsheet that reverse engineers times that would produce the right results in Practiscore, then enter them into Practiscore from the spreadsheet. The downside is, you can't enter that stage's scores until everybody has shot, because theoretically the very last shooter could turn in the best score of the match. At least you could post the hit numbers (if not the stage points) as you go along.

Having played around with scratch paper for the last half hour or so (I am a trained mathematician—don't try this at home! ), the best you can hope for with direct Practiscore entry is an approximation of the results you want, and it would require multiple different penalty classes. For example (just picking these out of a hat), say I have 40 targets. I start by giving everybody a 25 second raw time. For each shooter, count misses (hit all 40, get zero misses). I would need to do something like have the first fifteen misses be 1 second penalties, the next ten misses be 2 second penalties, the next five misses be 5 second penalties, and so on. And the more misses, the higher the penalties have to be to push the low scores down towards zero.

That would be its own scoring headache, obviously, and even that system gets less accurate as the top shooter moves farther away from shooting clean.

So, I recommend score it manually and put in a feature request to Practiscore to support fixed time stages in Time Plus (Points) format. It's already in their USPSA arsenal, so hopefully it isn't too hard for them to add it here...

If you want to talk scoring approaches in more details (like, with formulas and such), feel free to PM me.

[Bryan 45]

Jomar, I like the idea of the stage. I say just make the scoring easy and if the points don't work out linear, at least everyone will have fun shooting the stage. No need to try to reinvent the wheel just to make it "fair". If you have shooters that are shooting the match for some other reason than just for fun, tell them thanks for coming and they're welcome to shoot somewhere else next year.

[J.Schmitt]

The stage idea is a good one, and we did it last year and it worked well. The results are http://mn3gungroup.org/wordpress/wp-content/uploads/2013/03/2013_TSC_Final_Scores_-_TACTICAL.pdf and you are looking for STAGE 7.

[Jakobi]

What if you ignore the par time (for scoring purposes) and assign each hit target a bonus time value? For example, Shooter A hits 25 targets within the par time. At -2 seconds per target the shooters time is -50. Shooter B hits 16 targets within the par time for a score of -32. Shooter A's score is -50/-50x100=100 and Shooter B's score is -32/-50x100=64. Shooter C hits no targets and scores 0/-50x100=0.

-Jake

[Benelli Chick]

What if you ignore the par time (for scoring purposes) and assign each hit target a bonus time value? For example, Shooter A hits 25 targets within the par time. At -2 seconds per target the shooters time is -50. Shooter B hits 16 targets within the par time for a score of -32. Shooter A's score is -50/-50x100=100 and Shooter B's score is -32/-50x100=64. Shooter C hits no targets and scores 0/-50x100=0.

-Jake

That should work as long as it takes negatives...because a negative divided by a negative is a positive!

Nice thinking!!

[ziebart]

It sort of works in Android ver 1.2.18. I say sort of because you still have to enter a time, but you can have negative times after bonuses. Additionally due to problems with negative times in normal stages with bonuses the algorithm skews the math.

I plugged it into a time plus points match with each bonus being 1 second each.

Shooter 1: raw time 1 sec, 11 bonus, time -10, stage pts 100

Shooter 2: raw time 1 sec, 6 bonus, time - 5, stage pts 88.89

Shooter 2: raw time 1 sec, 0 bonus, time 1, stage pts 78.43

This seems to weigh scores to the top.

Currently I think the best solution is similar to what was already posted. Build a second match in a device dedicated to that stage using the USPSA fixed time function. Then use the stage percents from that as the times to input into the main match. This would have to happen after all shooters have finished that stage, but I think data entry could happen quickly if a couple people helped. Also a stage like this is likely to finish before the others.

One issue I have is getting scores to show complete in the USPSA fixed time if the shooter hits all the targets. It does not seem to want to generate a HF. If I add a penalty, save scores, and then remove it by using the edit function it will generate a score. I guess not too many people shoot perfect fixed time stages, or I am doing something wrong.

Edited June 30, 2014 by ziebart