Duck boats.net

[don novicki]

I was at DuckBoats.net and read the harvest numbers for all the states, and hunter participation numbers. For Pa. 2022 vs 2023. If you can believe it, which I don't, Pa picked up an additional almost 9000 duck hunters and Pa managed to shoot almost 8000 more mallards than in 2022. In addition we supposedly shot more than double the number of Canada Geese than in 2022 and doubled the number of goose hunters that we recorded in 2022. I'm throwing the B.S.flag on this. The report also indicates that there were ZERO Redheads and some other species taken in 2023. How do they know this stuff? We do not fill out a day to day report of species killed, and then send it in to the powers that be. This whole report look suspect.

When you look at SC, which is supposed to be one of the worst states for waterfowl, you'll be blown away at the numbers.

I don't believe a word of it.

Maybe someone could post the report here. I have very limited computer skills.

[SpinnerMan]

This is very long, but if interested, I hope this explains how these work.

How do they know this stuff?

This is what statisticians do. It's not really that challenging of science.

Last year I sent in the wings or for geese the three outermost wing feathers and the tail feathers. Along with the feathers, you provide when and what county it was shot. From that they can tell what species of duck, what sex of duck, and adult or juvenile. From the goose they can tell adult or juvenile. Plus they get information on distribution. They've been doing this for decades. This provides them one set of year over year data points.

You can see how given the same number of hunters sending in wings and tails that they can estimate how the harvest rate of each species changed year over year. So if they got twice as many geese, then the mean estimate would be that every hunter got twice as many on average. Of course, you can also see that there is a margin for error on this estimate.

It's like flipping a coin without knowing how many sides are on it and estimating what those sides are. Flip it a few times, you have a pretty good chance that you will be far from 50/50 which we know is the exact answer. However, flip it 50 times or 100 times, the probability from being significantly different from 50/50 is very small. That's the basic underlying theory and the math is actually pretty straightforward for all of this.

What you worry about is that those willing to collect and turn in wings are not representative of the average hunter. Maybe guys that kill a lot of birds every year do not want to be bothered. So the people that shoot the most birds are underrepresented in the sample. There are ways to check for this and correct for this. We all fill out the HIP surveys ever year. So there is a whole bunch of data to check against the wing samples. Do the collects have the same distribution as the HIP surveys?

There are other ways to get data. Places where you have to report your kill every day when you check out. There's more data with a long history. Did they shoot double the geese at those places? This will give you more or less confidence in your data.

You suck up as much data as you can and from that you can make pretty good estimates of things that happen frequently meaning the error bars are pretty small compared to the average. We shoot a lot of geese every year, so it's hard to see how they could be that far off.

However estimating things that happen infrequently is much harder.

The report also indicates that there were ZERO Redheads and some other species taken in 2023.

What this really means is that they did not see any in the wing surveys or any in any of the log sheets which doesn't mean absolutely zero. It means that it was less than a small number with a certain probability. Depending on the amount of data collected, it would mean the value calculated is zero and there is a 95% probability that the actual value is less than 50 or 20 or whatever the math works out to be. The more data they have. The lower that value will be.

That's the basics of how it works. The math is really easy. The really hard part is taking the data that you have collected and adjusting it for the biases. This means that the population of hunters that you are measuring probably doesn't have the same characteristics as the population as a whole. Your sample might underrepresent those that hunt a lot or those that hunt a little or those from different parts of the state are under or over represented.

This is no different than political polling. The math is the easy part. The hard part is converting the numbers from those people that answer the phone and are willing to answer your questions to what the actually votes would be if the election was held on that day. This is a likely voter survey. Registered voter surveys are NOT trying to predict the outcome and tend to be biased in favor of the Democrats. The likely voter surveys also assumes that the pollster is trying to be the best scientist that they can be and not simply trying to convince the other side that it is hopeless, so don't bother to vote. Even for those that are absolutely trying to be the best scientists they can be and are searching for God's honest truth, they bring their own personal biases which influences how they process the information. If they really want Kamala to win, they will likely error in her direction and the same for whatever they are polling. It's not intentional. It's just how us humans see things.

I'll give another example of how bias works. My mother was the score keeper for a large fraction of my baseball games from the time I was 9 until I quit playing competitive ball after 15. Her bias was that she did not want to be seen as being unfair in favor of her kids. Too many mom's would be score keeper and their kid would hit a slow bouncer to the pitcher, the pitcher would drop it, kick it a couple times, pick it up and throw it over the first basesman's head. Then their mom would come over the PA "Single." My mom cost me two batting titles. My season when I was 15. I batted about .600. We played about 12 games. I was safe on error 10 times I don't think anyone else on the team was safe on error more than 3 times and most 1 or 2 all season. That's where problems can come in. If they errors are all one direction, then you can get BS. If the errors are unbiased, my mom maybe scores one of my errors a hit and one of my hits an error and the average works out. But with at most 4 and more likely 1 or 2, the average was much lower than the actual average. Correcting for the biases, meaning assuming I was the average of the rest of the team and had 2 safe on error and 8 more hits than recorded would mean my actual average was over 150 points higher. That's the kind of thing you have to do to correct for biases, but you bring your own biases in when you are making the assumptions needed to correct for those biases. But for the waterfowl numbers with tons of good data, its hard to see how any big changes are not real. Maybe it was only 80% more or maybe 120% more, but its not flat.

On a personal note, my mom and dad were by far the most respected people in our little league at that time because every kid knew they were getting a fair shake and that they weren't doing things to unfairly help their kids like so many parents did intentionally or unintentionally. Years later I would run into kids and they would ask about my parents. They really did impact kids lives for the better. I wasn't at all mad at my mom. I never said anything until decades later and it was just telling a story like this. She was mortified, but I told her it was a good thing. That's why I didn't say anything. It will be 7 years this fall since my parents past just 8 days apart and way too young. I miss them, especially during hunting season.

[don novicki]

Thanks for the detailed explanation. It makes sense now. You batted 600 when you were 15. Dam, now that's impressive........

[SpinnerMan]

Thanks for the detailed explanation. It makes sense now. You batted 600 when you were 15. Dam, now that's impressive........

600 with 10 safe on errors in about 12 games

I was good enough to make bad decisions about college. Thankfully our high school coach was awful. I quit after my freshman year. I got 2 at bats all season. One I hit about 400 feet or better on a field with no fence. It was late in the season and I had already quit in spirit. So I loafed my way to second for a double. Kids remembered that smash years later.