New Mexico Big Game Draw Simulator

[Bigeaux20]

To be more helpful here is an example.

Unless this has changed, when I last looked I could not determine if New Mexico runs one draw that sequences resident, non-resident, and outfitted hunters, or if it is two or even three separate draws. The way the results shake out to me I have to conclude it's one draw and everyone is pooled and sequence is set and tags are awarded, and once outfitted hunters and non-residents hit their maximum allotment they are ignored. But I don't know that for fact. Your simulator would have to know that for fact to be accurate. You would also have to know how many people apply in all three pools.

You may be correct and I don't think we are far apart. The way it's laid out as I understand it is everyone goes into one applicant pool (res, non-res, outfitter). They random sequence all the applicants, and go to the first choice. (certainly the first guy gets his first choice). Throughout the process it continues. If there is a first choice quota available for the first option, it's awarded. If not, they look at the second choice, if there is quota available (once again only pulling from the appropriate pool), it's awarded, if not, third choice, same process. In essence all hunters are looked at, but if the quota for the hunts you have selected is exhausted, you are "drawing dead". Each applicant is evaluated so to speak, but it's actually quite high before every tag is gone for some units. Some, it's rather quick. It was enlightening to me seeing the numbers where hunts are exhausted typically. It's impossible to replicate human choices and swings, but I think I am working from the same data as everyone else and I think I built the trending from the previous years into the model as well. Good stuff though. Appreciate it.

 

[Bigeaux20]

GoHUNT as well as at least one other guy use the actual nm draw database for their simulations.

Yes, and this is the same data I used as well. Everyone can see the data on the website.

 

[Bigeaux20]

Sometimes the old ways are still the best.

View attachment 852121

I wouldn't bet money that is more or less effective as anything else.

 

[Dos Perros]

Do they even publish data for which tags are applied for first second third etc?

 

[Bigeaux20]

Does anyone want to offer a set of choices to run through the deal, and see what it spits out. Any takers?

 

[Bigeaux20]

Do they even publish data for which tags are applied for first second third etc?

The publish how many applicants for first, second, and third, as well as tags that are awarded s first, second and third choices across the board for resident, non-resident, and outfitters. None of it's a secret someone has special access to unless they are somehow getting the actual results and can see at thaw sequence order tags are issued. That's the part I "think" I have replicated and is a different way of looking at things. Once again, I don't know for sure this is of any value, just opening up the debate as to whether it is.

 

[Maverick1]

The state of NM has a spreadsheet with the draw odds on their website.

Take a look at it. Any of the remotely decent tags are all low single digit odds.

Creating a machine that will just tell you your odds are really low, exceptionally low, or ridiculously low is just an exercises in mental masturbation, IMO.

But you seem to be fired up about it, so have at it.

Good luck in the draw!

 

[wapitibob]

Yes, and this is the same data I used as well. Everyone can see the data on the website.

Really? I applied for VV 1st choice, what were my 2nd and 3rd choices

GoHunt uses the draw database, nowhere near the same data you are using. In the end, they're last years odds anyway and we all know the results.

 

[Jakeb]

I think you are putting a lot of brain power into a draw that basically you’re probably not going to draw. I put in for New Mexico every year as just a bonus tag. I remember I used to stress about having to apply in CO before the NM results were out worrying about drawing 2 tags. How foolish I was lol.

Odds are low, knowing just how low doesn’t seem to make much of a difference. New Mexico has no secrets as far as non-res elk hunting goes. Unless you want to bow hunt during Christmas. If the unit has any elk at all the odds are trash.

 

[fatlander]

Hey, you’re hitting right where I’m at. I’m putting in either way—gotta play to win, right? But how do I make sure I end up somewhere that’s worth the trip? That’s the itch I’m scratching. All this data floating around—ONX, GoHunt, the works—it’s gotta be worth something, I’d think. So how do we crack it open? How do we sift through it and figure out if there’s more to pull out—something that actually moves the needle? You tell me—what’s your gut saying about how we turn that pile of numbers into a real edge?

Honestly, I just don’t personally think there’s but a handful of tags in New Mexico that justify the long odds that every tag in New Mexico has. Not sure how to get an edge on that, because everyone knows which tags I’m talking about.


Sent from my iPhone using Tapatalk

 

[Jethro]

Hey, you’re hitting right where I’m at. I’m putting in either way—gotta play to win, right? But how do I make sure I end up somewhere that’s worth the trip? That’s the itch I’m scratching. All this data floating around—ONX, GoHunt, the works—it’s gotta be worth something, I’d think. So how do we crack it open? How do we sift through it and figure out if there’s more to pull out—something that actually moves the needle? You tell me—what’s your gut saying about how we turn that pile of numbers into a real edge?

No math is going to move the needle in NM. Odds just too low and there is no secret that will pop out from running the numbers. Sorry.

Divide your hunt choices into 3 categories. Like winning the power ball, have to get unbelievably lucky, and going to need some luck. Pick 1 hunt from each of those categories and apply them 1,2,3.

 

[SDHNTR]

Keep it to yourself. People should have to work for it. There’s already too much out there making western hunting too easy.

 

[Legend]

Alright, let’s slow this down for a minute and talk about something that’s been rattling around in my head. Picture this: I’m just a flatlander, born and raised east of the Mississippi. Where I come from, the ground’s not just flat—it’s soft, forgiving. Nothing like the rugged terrain you folks out west deal with. Now, I’ve been trying to wrap my head around this big game draw system—trying to crack the code, you know? And me, being a little bit of a numbers guy—okay, maybe a numbers nerd if I’m being honest—I started wondering: what if there’s a better way to get a feel for this?

They say experience is the best teacher, right? But who’s got the time—or the tags—to rack up decades of draw experience? So I thought, why not create a way to fast-track that? I'm not a program writer or coder or whatever you call it. I'm just a guy who spends too much time obsessing about this stuff. What if I could run a simulation—multiple draws, over and over—until the patterns start showing themselves and somehow make sense of what these "draw odds" really mean? That’s where this idea came from. My approach was if I made these three selections, and simulated the draw, how many times would I win. Here's something I didn't expect. If the odds are 4%, the simulation results weren't 4 times out of 100. It's astonishing.

I took past draw data—real numbers, real outcomes—and built something that creates a random pool of applicants, just like the real thing and with the same strategies and tendencies. Then it adds your choices into the mix and runs the draw as many times as you want. One simulation? Sure. Ten? Why not. A hundred? A thousand? Now we’re talking.
And here’s the kicker—it’s built to mimic how the New Mexico Big Game draw actually works. Random sequence, choice-based, the whole deal. And you can run it for any animal in the draw, any hunt code, any weapon, and any category of resident, non-resident, or outfitter. In the end it's spits out a spreadsheet that gives some interesting information. I’ve got some examples here, some screenshots to show you what it looks like in action. I did all this as a personal tool, but I’m even kicking around the idea of putting this online, maybe a website where anyone can use it. But here’s where I need you—I’m not too proud to ask for help. What do you think? Is this something that’d be useful to you? Something worth digging into deeper? I’m all ears—lay it on me. Is there something here? Pick it apart. What makes you interested or skeptical?

I am not a math modeler. However I have worked with a few and even had them model some draw scenarios.

I believe the starting point would be a Monte Carlo model. Customized and then calibrated to real historic data. From here you can run scenarios.....millions of scenarios. This is the way! I think.

The draw odds sold by the big companies nies for NM and Nevada are just wrong. So wrong there may not be much applicability.