(The thread has gone a million directions since I started this, but I still want to post it because this is the internet and no one can stop me.)
39% of statistics are made up on the spot.
Are you referring to my 90 - 93% estimate?
It's just an estimation but its probably not far off, its not important to be precise to the second decimal point here
I think that's a joke that I get the honor of explaining!
It is implied that the '39%' part was made up on the spot by Deadlock, which he did for comedic effect.
Its a joke used to insult people who throw figures around without referring to actual recorded data.. I did "make up" the percentage but that is based off of memory, which is good enough for our present purposes
I was indeed poking some fun a people who use made up statistics. It was not directed only at you, and I did not intend it to be insulting. Given the direction of the thread, it is understandable that jokes could be taken in that way, and I am sorry I offended you. I do strongly agree with the sentiment the joke conveys. When people throw statistics around in these forums, they are relavant to the conversation less than 10% of the time and and have less than a %1 chance of being accurate. (...see what I did there)
You claim that your estimate based on memory is "good enough for our present purposes". While "good enough" is totally subjective, your number isn't supported terribly well by your recently logged games; however, the sample size is probably too small to make a meaningful percentage claim from. Based on the 37 pro games that included King's Court that you have played on the new system, in which you have skipped it 7 times, there is a less than 3% chance that you think King's Court is skippable 7-10% of the time. To get a ~95% confidence interval I would need to say that you skip (and presumably, think you should skip) King's Court 9-33% of the time (unskippable 67-91%)
Human memory is an incredibly bad source to base statistics on. This is not only because of the fact that it can't reliably hold the quantity of trials neccesary to generate useful statistics, but also because it is just downright bad at remembering these types of things.
Now, I did some math (probably incorrectly), and it was fun, but, even if the data was better, and my math and application of it was perfect, I consider it all totally irrelavent. Maybe it helps support my opinion that made up statistics are not worth using to defend an argument, but we need a well defined argument to start with before we attempt to support it with anything. I do feel that the response you have received in this thread is in part due to the way you positioned, stated, and defended your argument. This has been touched on by a couple posters while I have been very slowly finding time to write this.
The biggest issue I have with your position ("KC is far more skippable than people think.") is that it is incredibly hard to define. The primary reason is that what "people think" is not a single position. There is a general concensus on the forum that KC is "rarely skippable", but in my opinion, getting more specific than that is not defesable. What "rarely skippable" means to f.ds as a percentage is impossible to nail down because everyone can have a different picutre in their head as to what it looks like.
Another issue I see is that the position posed is very combatitive. Presumably, everyone who reads your assertion is a person, and considers themselves part of this community. To the reader, your position becomes "I think KC is far more skippable than you do". You can argue that this is pedantic picking apart of wording, but the emotion your position evokes is important. After reading your discussion in this thread, I would estimate, that I consider KC approximately the same level of skippable as you do, but I don't agree with your claimed position.
The last comment I would make is that the position is execptionally difficult to defend. Even if we had a definitive position from the community that KC was unskippable on 82% of boards, attempting to defend an arguement that the number is too high is an exceptionally tall order, and a single game of Dominion is barely going to scratch the surface.
Since I have criticized the argument as written, I feel obligated to give my opinion on a better way to state and defend this position. A less combatitive position could be
"KC can be skipped more often than some think" For this discussion I personally would have gone with a questioning title such as
"What type of boards make KC skippable?" instead of a claim statement. For the post itself, I would defend the position in this way:
"This is a board I played recently where I believe the general concensus would say to go for King's Court. I think that it should be skipped due to the presence of strong junking without trashing despite the presence of +buy. I was able to beat my opponent decisively by ignoring King's Court and pursuing a money-based strategy. Do you think King's Court should be skipped on this board?"I think we could stand to be more friendly around here when debating topics like this even if we think the person we are arguing with is an idiot. I think a lot of arguments get heated based on a difference in opinion that is effectively non-existent. We see in this thread, that the OP's position wasn't even on the side of the stats from top 20 players that he thought it was. I suspect the "100% of games are engines" caracature of reality that is pervasive around here and the attitudes that go with it shaped his argument.