CO moose draw stats

[COLOelkman]

I also posted this in the Moose, Sheep Goat page.

Below are last years odds for residents to draw a moose tag (bull and cow combined). They do not separate out the numbers for bull tags but generally speaking, odds are a lot better to draw a cow tag than a bull tag.

2018 CO Resident Moose draw odds via CPW website
Weighted Res Res Draw/Apps
W Ppts Apps Draw (% odds)
17 608 38 6.25
16 368 10 2.72
15 516 14 2.71
14 474 10 2.11
13 499 12 2.40
12 645 19 2.95
11 538 12 2.23
10 618 10 1.62
9 689 13 1.89
8 697 25 3.59
7 774 20 2.58
6 997 30 3.01
5 1024 27 2.64
4 1164 21 1.80
3 1318 29 2.20
2 1621 25 1.54
1 1984 46 2.32
0 2460 35 1.42

So here's my question/problem with the current moose draw system. Next year, they've added a $50 fee to apply for moose. I have a lot of moose points but there is really no better odds if you have high points or low points. I discussed the moose odds with a CPW officer at the Denver ISE and he claimed that those with high points have a "significantly" greater chance to draw. As my instincts told me that was not true, the numbers confirm my suspicion that points don't really matter. You kinda hafta understand the draw system a little to understand why. So, trying to decide whether to now shell out $50 every year for about a 2% chance to draw (prolly much lower for a bull tag). I plan to at least comment to the CPW on their current system to see if they might consider modifying it some to provide a little more favor to those that have been applying for many years. Any comments?

 

[DW]

Per your post, max points was at least twice as likely to draw as the majority of other point levels.



#livelikezac

 

[nripepi]

I think it was likely just an anomaly that 17 had the highest percent last year, I bet most years it is not the case when in fact it probably should be the case. I agree with the OP in that the system is a bit backward in that it divides your random number by your bonus points to get your final number instead of giving you X number of random numbers (chances) per the number of bonus points you have. It is probably an advantage to have 17 bonus points versus 1, but it is not even close to a 17:1 advantage. What would be interesting to know is how many of the lowest random numbers drawn get the tag or don't? That will tell you how much it matters.

 

[Coloradoboy]

I'm 3 plus 11 going into it this year. Maybe there's an end in sight....

Coloradoboy

 

[DW]

I'm 3+15, and after watchin my buddy draw with 3+1 I dont believe there is an end.




#livelikezac

 

[COLOelkman]

Looking at the numbers, you see way more people draw that are in the lower half of those eligible. 35 people drew with 0 pts and 43 drew with 1 point. Granted there are more people in the pool which tells you it's almost a random draw, which is why the numbers are relatively equally distributed. I pay for points in other states with very low odds but with their systems at least you will eventually draw if you stick it out and get in the top points tier. Not so with our moose system. I plan to go to the CPW office and try to find someone in the know to submit a comment as the online comment area doesn't seem to have a place for drawing comments/questions.

 

[DW]

It's why you should get your name in the hat for each weighted point. Lower point holders could still pull the random number that gets them the tag and higher point holders get a slight advantage for applying for years. As it is now everybody gets one chance at a low random number. It is a random draw.



#livelikezac

 

[COLOelkman]

I agree with you DW! I'd also favor giving 1 random number for each weighted PP which would be a more fair way to do it. Everybody would still have a chance but it would award those that have invested more years in the draw. I have a friend that has drawn 3 moose tags so maybe I should have him fill out may application as he's one lucky dude!!!

 

[DW]

Yep that's what I meant by name in the hat, 15 weighted points=15 random numbers=name in the hat 15 times. As it is the guy with 0 weighted has got just as good a chance of drawing as I do with 15 should he get a low random number.


#livelikezac

 

[JRABQ]

LAST EDITED ON Jan-31-19 AT 12:54PM (MST)[p]LAST EDITED ON Jan-31-19 AT 12:42?PM (MST)

I'm a bit surprised by your results. A guy on bowsite in 2017 did an analysis for Mtn Goat and showed that weighted points did in fact help your draw odds, pretty much in line with the "extra tickets in the hat" (or bonus point) expectation.

[https://forums.bowsite.com/tf/bgforums/thread.cfm?threadid=463226&messages=133&forum=13]

Of course the thread devolved into a bunch of arguing and confusion.

I would guess for the moose results the data might be skewed because low point folks are going to put in for easier to draw hunts (including cows), and folks with more points are holding out for more premium tags. But that bias (if it exists) doesn't seem to show up in the Mtn. Goat data in the link. So that is puzzling.

edit; going back and looking at the bowsite thread, the author (sticksender) did strip out nanny and population reduction hunts from the Mtn. Goat data set. Don't know how big of an effect that was.

 

[elks96]

Should just be random draw. No points needed and everyone has equally opportunity.

The truth is even with points it is still pretty random and it should be, no one should have any preference.

 

[COLOelkman]

LAST EDITED ON Jan-31-19 AT 03:53PM (MST)[p]I think it's kinda too late to go random as once they start points, it's a slippery slope that they can't go back on. Odds are always gonna be low no matter what but at least the guys at the top should get a little more preference than the current system. Most States draws do this and I think the CO moose is one that is very different.
JRABQ - I'm not up on the sheep or goat process so I'm not sure if they are the same system or not. Regarding putting in for better hunts, I don't believe that really applies to moose. A few years back I ran the odds for every unit (bulls only) and there was very little difference in the odds so it really didn't matter much which unit you picked. Some differences sure but basically negligible. Not sure if thats true on the cow moose hunts.

 

[JRABQ]

Moose, sheep, and goats all use weighted points.

 

[DW]

I dont know how you could make it any more random than it is. You couldnt #$%# high point holders anymore than you already have. Well unless you decided to eliminate points and go straight random from zero. FFS.


#livelikezac

 

[Waygoner]

I have a question about the application process. Once you hit 3 points, can you apply without buying a preference point? Meaning, can I just stay at 3 points and still apply every year?

 

[JRABQ]

That is what I recall reading for this year, for now paying the extra $50(res) or $100(non-res) will be optional for moose/sheep/goat.

 

[COLOelkman]

If that's true and looking at the numbers, it seems like the best strategy would be to not apply for any more points, save your $50 and basically have the same odds.

 

[JRABQ]

>LAST EDITED ON Jan-31-19
>AT 03:53?PM (MST)
>A few years back I ran the
>odds for every unit (bulls
>only) and there was very
>little difference in the odds
>so it really didn't matter
>much which unit you picked.
> Some differences sure but
>basically negligible.

Well you got my curiosity up and I spent a few hours looking at the resident Bull moose only data (no cow or either sex hunts). Looks to me that there is a clear unambiguous trend in the 2018 data that odds increase with weighted points, which was not seen in your total data set including cows, etc. This trend also agrees with the Bowsite guy's goat analysis from 2017. Yes there is still some statistical "noise" in the data, which is expected due to the small sample sizes, but the results look just like a "bonus point" scheme (extra tickets in the hat). Some might even argue that the highest point holders (WP=17) draw at a higher rate than expected, but I doubt the difference is statistically meaningful.

(Minor point; I did not include the youth numbers because they were in separate columns and would require more work to add them back in; the youth numbers are really small and should not affect the outcome.)

 

[jims]

Everyone seems to have their own opinion on which system is the fairest and if those that apply longer should have a better chance of drawing tags. With premium limited units in such high demand things get controversial in a hurry!

I kind of like the idea that those that pay $50/year to apply and devote years and years to applying in a draw system to stand a little better chance of drawing high demand tags than those just starting out. I have a difficult time throwing $ to the wind with almost 0 odds of ever drawing a tag without some sort of pt draw system. Look at the current draw odds for desert sheep in Colo where there is no draw pt system! I like the idea that everyone has a chance to draw the same tag in any given year but those that have been in the game longer statistically have somewhat better chances to draw.

Some states (UT and WY nonres as examples) offer 1 tag to those with highest pts and a separate set of tags in a random draw. Some Wyo residents have actually complained because they have no point system for deer, elk, and antelope. Some guys may draw back to back to back tags while their neighbor draws 0 tags. That's another negative to not having some sort of point system! Another benefit is you pretty much can plan out hunts in Colo. You can pretty much predict which tag you will draw in any given year...with the exception of super high demand tags like moose.

Looking at your graph it doesn't look like you ran any stats to show if 0.22% draw odds with 0 pref pts is any different than 6% draw odds with max pts? The max pts guys jump from 3.46 to 6% odds which is a little weird. I doubt if there is any significant difference in draw odds between those apply with 0 and close to max pts? I'm also not sure if looking at this graph it shows exactly what goes on in Colo since there is a random....low assigned number deal that seems mighty complex and complicated!

 

[JRABQ]

Answer to jims;

>Looking at your graph it doesn't look like you ran any
>stats to show if 0.22% draw odds with 0 pref
>pts is any different than 6% draw odds with max pts?

OK. Might not be what you where asking for, but if you assume the draw odds are independent of your weighted points you get a mean of 2.19% and a

standard deviation = 1.44%, max deviation = 3.81%

If you assume a bonus point (linear) equation with only one adjustable parameter you get

standard deviation = 0.72%, max deviation of 1.84%

So the linear expression yields much better stats than assuming a constant value, using all of the data.


>The max pts guys jump from 3.46 to 6% odds which is a
>little weird.

Yes it is. There are many anomalies in the data set if you just compare one point to another. People with 12 points had better odds than those with 14. Do I really think that is the case; No, I think it is just an artifact of the poor statistics inherent in a small sample size.

You can overcome this issue to some extent by testing models using the ENTIRE data set. The results above showed that a linear expression was a better representation of the data than assuming a constant value.

>I doubt if there is any significant difference in draw >odds between those apply with 0 and close to max pts?

OK, the standard deviation values of the two possibilities tested says otherwise, in my opinion.

I will also see if I can add some uncertainties to the data on the graph.


>I'm also not sure if looking at this graph it
>shows exactly what goes on in Colo since there is
>a random....low assigned number deal that seems mighty complex and complicated!

That is true, the graph and analysis say nothing about the crazy scheme CO uses to assign a "random" number. That is another discussion.

 

[COLOelkman]

Does anyone know what the parameters are for their "random" number?The CPW guy told me that he thought it was something like a 5 digit number which he said makes it more favorable for people with high points. Not sure I understood the logic of that as if everyone gets a 5 digit number, that's quite a bit different than a number from 1-5 digits. Seems like if it's 1-5 digits, it would then benefit those with higher points but not if everyone gets a 5 digit number. Anyway, I plan to go to the CPW office Mon or Tues to see if I can find anyone that can explain it any better.

 

[DW]

#livelikezac

 

[Horniac]

https://cpw.state.co.us/Documents/Hunting/BigGame/Weighted-Preference-Points.pdf


Horniac

 

[COLOelkman]

Thanks for posting the method for this process as I know understand how it works. So doesn't it seem that if you get lucky and end up with a really low number after all the conversions (which is basically a random number), you'll likely get a tag regardless of your points. Points do factor in but not as dramatically as if you had the same number of chances as your weighted points (+1). Just my opinion but it doesn't favor the higher points people as much as I'd expect or think. I understand odds won't be great with any system since there are so many applicants for so few tags but seems to give people with very few points a much better chance to draw a tag than I'd think.

 

[DW]

LAST EDITED ON Feb-01-19 AT 10:38PM (MST)[p]
Because the random number they use is such a large number I agree it doesn't matter how many weighted points you have. It's all in the luck of the random draw number. Which is why I say it more or less is a random draw

Using the example in horniacs PDF, even if josh had 1000 weighted points, kate would still pull the tag with 2 weighted points because she drew a lower random number.


#livelikezac

 

[jims]

LAST EDITED ON Feb-01-19 AT 07:43PM (MST)[p]My guess is that JRABQ's numbers are the actually percentages of how many guys drew at each pref pt...even with the random number included. The odds of drawing a tag with 0 pref pt is close to 0 and even with 16 pref pts the odds are still pretty darn close to 0!

I really don't get too terribly excited about draw odds that are less than 20% or 1 in 5 chance! In 5 years of applying you have a pretty good chance to draw if draw odds are 20%. With a 5% chance it would likely take over a lifetime to draw if no one new starts applying for tags! The frustrating thing is there will likely be twice as many guys applying for the same tags 10 or 15 years from now....in another 20 years there may be 4 or 5 times more guys applying that now!

Another consideration is how many moose tags will be available 10, 20, or 30 years from now? In Wyo's case the grizz/wolf have put a major dent in the moose population in the NW corner of Wyo. Fewer moose + fewer tags! It's hard to say what may happen in Colo....right now the moose population is booming but what about unexpected circumstances such as disease, intro of wolves, etc....you never know!

With no point system the draw odds just get worse and worse every year because there are fewer guys drawing than new guys just starting out applying for tags!