For those who think Trout over miggy

[jalopy]

I said that if they find the right mathematical formula... therefore your wise comment that it may not be statistically valid is irrelevant...

I certainly don't want to speak to TO's intent but I think he was referring to the fact that your formula might not be as good as you think it is considering the conclusion it is making.

 

[DragonfromTO]

I have explained it to you 5 times, if you dont want to read the entire post then thats ok, but your fixated on something that has very little importance... Not sure if you are trying to be an asshole or if you just are one!!

If you gave the kind of explanation that you've given here on a test in an actual statistics class how many points do you think you'd get for it? It's not that I didn't read your explanation, it's that your explanation doesn't actually explain anything.

 

[MilkSpiller22]

I certainly don't want to speak to TO's intent but I think he was referring to the fact that your formula might not be as good as you think it is considering the conclusion it is making.

never shared my formula... it is a good thing for u to assume that i dont know what i am doing... even if it is not good, how is that relevant... I have never said that i am right, all i said is that it makes sense to me... In fact i have stated that i need to tinker with some formulas still... so clearly i know it is not perfect... the only part dragon has a problem with, since i said it is that i use 3% whether the player makes the playoffs or not... he asks why... i even said that it was a little arbitrary, but it is the way i prefered it when i was making my own formulas...

 

[DragonfromTO]

Wow, that's pretty cool. Not really what I was expecting. But I see how DK could be of service. I like that.

Yeah it's loads of fun. It's pretty wild how much you can learn about players individual tendencies and the process of team building just by doing basic draft preparation and stuff too. Like if I told you I used Ronny Cedeno, Cody Ransom, Adam Rosales and Alex Gonzalez to cover shortstop last year you'd probably think that was a position of weakness for me. But they actually combined for 40 HRs in 764 PAs for my team, because I used them the "right" way and got them in the best matchups and situations that I could.

I'll send you a private message with the link to my league website if you want to take a look. It's really a lot of fun, I like it a lot more than I like fantasy/roto.

 

[DragonfromTO]

never shared my formula... it is a good thing for u to assume that i dont know what i am doing... even if it is not good, how is that relevant... I have never said that i am right, all i said is that it makes sense to me... In fact i have stated that i need to tinker with some formulas still... so clearly i know it is not perfect... the only part dragon has a problem with, since i said it is that i use 3% whether the player makes the playoffs or not... he asks why... i even said that it was a little arbitrary, but it is the way i prefered it when i was making my own formulas...

The burden is on you. You have to show us what you're doing (or at least give us some kind of outline) and why we should trust it. You don't have to walk us through every step and every calculation, but you have to give us a lot more than you are giving. You act like you're insulted by the assumption that what you're doing is invalid, but what do you expect the default position to be for something like this? You have to show that it is useful/correct and why, you can't expect people to assume that it is.

 

[MilkSpiller22]

The burden is on you. You have to show us what you're doing (or at least give us some kind of outline) and why we should trust it. You don't have to walk us through every step and every calculation, but you have to give us a lot more than you are giving. You act like you're insulted by the assumption that what you're doing is invalid, but what do you expect the default position to be for something like this? You have to show that it is useful/correct and why, you can't expect people to assume that it is.

Never said you have to trust it... read dont react!!!

 

[jalopy]

never shared my formula... it is a good thing for u to assume that i dont know what i am doing... even if it is not good, how is that relevant... I have never said that i am right, all i said is that it makes sense to me... In fact i have stated that i need to tinker with some formulas still... so clearly i know it is not perfect... the only part dragon has a problem with, since i said it is that i use 3% whether the player makes the playoffs or not... he asks why... i even said that it was a little arbitrary, but it is the way i prefered it when i was making my own formulas...

No, you haven't shared your formula but you have given us the results. Since I don't happen to agree with those results, I question the science behind it. Based on the portion you shared on the 3%, I have serious concern about the other 97%.There is no mathematical reasoning for including team success. That is why you are forced to make an arbitrary judgment.
I apologize if I am being overly critical but I take it as a personal affront when mathematical reasoning is mis used.
I laud your efforts and would love to see and discuss your formula further. With an open mind, discussing it with others can only benefit you.

 

[Darkstone42]

Never said you have to trust it... read dont react!!!

Okay, overall I haven't been terribly critical of you in this thread, but what kind of discussion statement is that? Read don't react? This is a discussion board. We talk about stuff. We don't read and not react. That's silly.

You're presenting some calculation here which you believe accurately assesses player value. Providing your end results without any comment on your method, even when prompted for it, is ridiculous. You said it's not perfect, and yet you won't let a massive community, many of whom live and breathe baseball, some significant background and/or experience in statistics and other mathematics fields which may be relevant, who may be able to provide valuable insights and bring up valid points of contention for your methods, both of which could improve them to the point you claim to want to get them. You'll have to excuse us for "reacting" and questioning your method (or, in fact, if you even have one at this point) and how well it truly assesses player value.

Come on, have a discussion. Stop being so guarded. You've completely defeated the purpose of posting your "findings" if you don't.

 

[Darkstone42]

Looks like we said the same thing, Jalopy. I was just a little meaner about it.

 

[jalopy]

Never said you have to trust it... read dont react!!!

Your own personal Pandora's Box.

 

[jalopy]

Looks like we said the same thing, Jalopy. I was just a little meaner about it.

Not really meaner. He has painted himself into a corner and realizes it.

If milk does indeed have a formula, I think it would be an interesting exercise to collectively try and improve it. If he has done his research, it would save a boatload of time rather than doing it from scratch. A list of assumptions is where I would start.

 

[Darkstone42]

Not really meaner. He has painted himself into a corner and realizes it.

If milk does indeed have a formula, I think it would be an interesting exercise to collectively try and improve it. If he has done his research, it would save a boatload of time rather than doing it from scratch. A list of assumptions is where I would start.

Empty or cherry-picking arguments do not sit well with me.

Then again, most of my reading has been peer-reviewed journals, and I get mad at those, too...

 

[jalopy]

Empty or cherry-picking arguments do not sit well with me.

Then again, most of my reading has been peer-reviewed journals, and I get mad at those, too...

Try sending your days trying to find an independent medical study. Sifting through biased conclusions to find imperial medical data is beyond frustrating. It is akin to peeling back the layers of an onion only to find you can't get through it all because of the tears.

 

[MilkSpiller22]

Ok... if you guys demand me to explain it my formula... wasnt trying to get too technical but in a short summary

2%-Games played

3% playoffs

5% position played(catchers do get a little extra here)

2% rank at position played

44%- formula using (RBI+Runs-HR)/PA and (RBI+Runs-HR)/total team runs scored)

44% using a version of RC27 that includes baserunning and other advanced statistics... also using a different version of total bases(not 1 point for single, 2 point for double and so on)


The percentages are all my preference...

The reason i use those 2 offensive stats is because i consider using Runs and RBIs as actual production of what a player DID while RC27 is the theoretical production where the team is irrelevant...

 

[Darkstone42]

Try sending your days trying to find an independent medical study. Sifting through biased conclusions to find imperial medical data is beyond frustrating. It is akin to peeling back the layers of an onion only to find you can't get through it all because of the tears.

I can't imagine. I work in the fundamental physical sciences, so there's a little less room for agendas to creep in, but points of ambiguity or "talking above the audience" really bugs me. Make it accessible so we can challenge you if we have to, but mostly so we can use the information you're presenting productively.

 

[da55bums]

They can't beat anyone when it counts.

The Tigers? or the A's? could apply to both

 

[Darkstone42]

Ok... if you guys demand me to explain it my formula... wasnt trying to get too technical but in a short summary

2%-Games played

3% playoffs

5% position played(catchers do get a little extra here)

2% rank at position played

44%- formula using (RBI+Runs-HR)/PA and (RBI+Runs-HR)/total team runs scored)

44% using a version of RC27 that includes baserunning and other advanced statistics... also using a different version of total bases(not 1 point for single, 2 point for double and so on)


The percentages are all my preference...

The reason i use those 2 offensive stats is because i consider using Runs and RBIs as actual production of what a player DID while RC27 is the theoretical production where the team is irrelevant...

I have three major concerns you may want to address:

1. Percentages are admittedly arbitrary. I recommend at least finding some historical or statistical trends to help you scale the parameters more scientifically. Using arbitrary ratios based on how you feel about it is akin to a scientist fitting data to a formula with three or four fit parameters. You can fit any result, and you don't get a true insight as a result.

2. What ratios do you actually use for weighting different types of hits in the last criterion? Are they the wOBA ratios? Those are based on league production that year, and could give you nice, justifiable numbers for scaling the hits.

3. I fear you may be "double counting" many offensive contributions, marginalizing defense even more than you actually intend to, and providing extra weight to certain types of hits than you intend to.

I hope this is helpful. I see some logic to the method, though, so it's got a good starting point.

 

[jalopy]

I am going to withhold comments until the morning. I really think that some assumptions would help clarify the direction of the formula. I see some serious flaws but I agree that the overall direction is worthwhile. I have some fresh chemo drugs running through so I will try to add some input in the morning.

 

[Darkstone42]

I am going to withhold comments until the morning. I really think that some assumptions would help clarify the direction of the formula. I see some serious flaws but I agree that the overall direction is worthwhile. I have some fresh chemo drugs running through so I will try to add some input in the morning.

I hope the treatment is going well for you.

 

[MilkSpiller22]

I do have a backround in statistics... one thing i have learned is sometimes simple is best... so instead of going historical, i just made the percentages as simple as possible...

My slugging percentage is ((BB-IBB)+1.75 1B +2.5 2B + 3.25 3B +4HR)/(Hits+BB)-still simple but going by .75 instead of 1 and i start with walks...

MY OBP is (H+BB-CS-GIDP)/PA

Where am i double counting??

Never claimed to be correct, but like i said it makes sense to me...