 12706 versus 12708. The minimum number is the same. That's because I deleted the data in the middle, so we still have the same bottom point. Quartile one is different, but not substantially different. The median, the middle of the data set is 124.19 versus 127, again different, but not hugely different. Quartile three, 136.8, 134.89, pretty close still. The maximum is the same because I deleted numbers in the middle, therefore there is the same maximum, but the standard deviation is now higher, right? So now I've got a 1301 versus 1166, which gives us an idea of the spread, which we might not have gotten a difference in the spread between when I'm looking at just these numbers up top. These might not have given me a sense as much as possibly a standard deviation of possibly more spread in the data and same with the variance, right? 135.97 versus 169.25, another kind of measure of the spread and that should give us an indication that there's something possibly more going on than maybe I would have picked up with just the first numbers up top with regard to the spread of the data. And then if I looked at those with the sample, these are just a comparison with the sample, they also being higher if I used the sample calculation instead of the population. And here's the actual histogram. So you can see what happened here. I deleted a lot of the data in the middle. So notice if I think about, well, what would have happened with my numbers, well, you would think the average would still possibly be pretty close to what it was because now you've got these two sides that still kind of average out to something that's pretty close to the mean. And then notice the minimum and the maximum are the same. So minimum, maximum, max are the same because these outliers weigh to the end. I didn't delete any of those. So that kind of makes sense. And then if you look at the quartiles, they're pretty close. And you would think maybe the quartiles would be substantially different if I did this in a small data set. But because the data set was so large, then I didn't really impact a whole lot of the quartiles if you just picked the number in the middle of the first, second, and third quartile, because even though I deleted a whole lot of data in the middle, we had a pretty significantly large data set. So you still end up with pretty close on the quartiles more than you might think. And but the standard deviation kind of does give us an idea. So you might say, well, these all look similar, but then the standard deviation does give us an idea that the spread looks significantly different as does the variance, you're like, okay. And so, and obviously that's represented here in the pictorial representation. Now just to wrap this up, if I did the same kind of calculation over here, I took my weights of my new data set where I deleted the numbers in the middle, I subtracted out the mean or middle point. This is the 127.06 now. Here's our differences. And the differences if I added them all up would add up to zero. I don't have all of them here because it was a very long data set, but it would still add up to zero with a new data set. And then if I square all of them, this is the squared amount. If I added up all the squared amounts, I would get to this and then I would take that and divide it by the count, which now I only have $19,999. $19,999 instead of I think it was 25,000 data sets before, right? So we removed a significant amount of the data, which is gonna be a variance of 169.25 and square root of that would be the 1301, 169.25 and 1301 is what we got here, 1301, 169.25.