What’s your go site to for stats?

why?

Mix and match?

what are your favorite stats for specific applications?

why?

Inspired by Brad

My post there:

WAR gets calculated slightly differently depending on your source.

I like Fangraphs. FWAR is the shorthand for Fangraphs Wins Above Replacement the link goes to the Fangraphs library where they define the stats.

For hitting I like wRC+ because it puts hitters relative to all other hitters (100 as average) by components of run creation.

For pitchers FIP is a great stat. It predicts ERA better than historic ERA does by calculating out if a pitched were to experience league average BABIP

For fielding I like OAA…
 

…afterwards I started thinking about if the shift and Manfredball might change my thoughts about my preference for WAR and if BWAR might overtake FWAR because BWAR weights fielding/pitchers to a greater extent.

For WAR and wRC+, I like Fangraphs.

If I'm in a hurry and just want a general view of a player or I want easy access to splits, I use B-Ref.

If I'm really digging into a player and want to know how each piece of their toolkit grades out, Baseball Savant.

B-Ref is usually my go to. I’m used to the UI and I like the easy access to splits stats. 

FanGraphs is good when I want to know velocities and pitch mixes. 

I truly don’t know enough about the WAR difference on either site to have an opinion. It’s just a quick way to determine value. 

I rarely go to Baseball Savant. They have a terrible interface on mobile devices. 

Time for me to dig in again. I was all old school until I read Law's "Smart Baseball" a few years ago. Taking it to Arizona to read again while I take in College Div I ball for a couple of months.
I lean on Baseball Reference - 1st, Fangraphs - 2nd. I use Fangraphs when I really want to dig in/drill down.

 

 

8 hours ago, Brock Beauchamp said:

For WAR and wRC+, I like Fangraphs.

If I'm in a hurry and just want a general view of a player or I want easy access to splits, I use B-Ref.

If I'm really digging into a player and want to know how each piece of their toolkit grades out, Baseball Savant.

Why wRC vs b-ref OPS+? The differences aren’t very big on the page

8 hours ago, Vanimal46 said:

B-Ref is usually my go to. I’m used to the UI and I like the easy access to splits stats. 

FanGraphs is good when I want to know velocities and pitch mixes. 

I truly don’t know enough about the WAR difference on either site to have an opinion. It’s just a quick way to determine value. 

I rarely go to Baseball Savant. They have a terrible interface on mobile devices. 

UI matters more than we talk about, but it’s very subjective.

agreed, baseball Savant is terrible on mobile.

FG’s glossary is a pain in the butt on mobile too. Adds cover it up and it drives me bonkers!

45 minutes ago, Richie the Rally Goat said:

Why wRC vs b-ref OPS+? The differences aren’t very big on the page

I think wRC+ does a slightly better job balancing offense but as you said, the difference between wRC+ and OPS+ is often so small it doesn’t matter.

I also like FG’s defensive metrics and options better than BR. I rarely use BR for anything if I’m looking into defense. 

I like BBRef for a quick bio or split stats. 

I'm old and old school...I do Baseball Reference. 

I'm not a huge stathead, but I do reference B-Ref quite a bit if I'm looking something up.  I do think I need to use Fangraphs more than I do though.

They've all got their strengths.

Fangraphs might be the most useful to me overall, not just stats, but depth charts and prospect scouting reports.  Their leaderboards are really flexible when you get the hang of them which is great if you're trying to look up something more comparative.  My main complaint is that the memory overhead seems to get pretty high which is noticeable for someone like me that typically has too many tabs open,

BBRef is more where I go for historical stuff. I also do like their minor league game logs and splits pages.

Baseball savant has great granular data though I think it will take some time before there are more useful aggregate level stats.  I think my favorite aspects is being able to get data on guys that are new to the majors.  Certain metrics stabilize pretty quickly and you can get some better impressions on how they are doing vs just the top level stats which are often deceptive early on.

I guess I'm mainly old school, I like digging into underlying conditions. But with overall comparison my  favorite stat is bWAR. But day to day stuff I like the splits w/ BB reference.

Fangraphs because I find their site easier to use. 

“The only nits I pick are that I like all numbers on a "relative to all other players" stat. A 3.75 FIP in 2023 meant a roughly average pitcher. In 2003, it meant a borderline all-star. 

Using FIP+ (or minus if you prefer) helps identify performance relative to the league at the time. 

OOA is a relative stat already, it just was not mentioned as such in your post. I think that is worth pointing out.”

 

@Minny505

i love this post and wanted to unpack this a bit.

you caught me in between understanding what happened historically and predicting future outcomes. Well done.

 

FIP predicts ERA but neglects context.

FIP+ provides the relative context but isn’t very predictive. (I always prefer the plus (minus seems counterintuitive to me)

how do you pull together both the context and predictive without writing massive paragraphs? I’m a lazy typist. 
 

 

 

1 hour ago, Richie the Rally Goat said:

“The only nits I pick are that I like all numbers on a "relative to all other players" stat. A 3.75 FIP in 2023 meant a roughly average pitcher. In 2003, it meant a borderline all-star. 

Using FIP+ (or minus if you prefer) helps identify performance relative to the league at the time. 

OOA is a relative stat already, it just was not mentioned as such in your post. I think that is worth pointing out.”

 

@Minny505

i love this post and wanted to unpack this a bit.

you caught me in between understanding what happened historically and predicting future outcomes. Well done.

 

FIP predicts ERA but neglects context.

FIP+ provides the relative context but isn’t very predictive. (I always prefer the plus (minus seems counterintuitive to me)

how do you pull together both the context and predictive without writing massive paragraphs? I’m a lazy typist. 

😂 And this is the conundrum of TV and radio broadcast teams across North America.

That said, I think FIP+ is both contextual and predictive in that it predicts the future performance of a pitcher relative to the league. A FIP+ of 110 predicts that a pitcher will be roughly 10% better than the average pitcher.

Tangent time: One stat I always wish was available readily available is something spFIP+, which was relative to only SPs, and it's counterpart, rpFIP+. For that mater 2BwRC+ for relativity on offense to to positional average. A 90 wRC+ from a catcher is roughly equivalent to a 105 wRC+ for a RF. Either a separate stat or something like the positional average in parenthesis next to the + stat. 

Maybe it's time to dust off the SQL skills...but I wouldn't even know where to get the data sets to begin. 

And FWIW, I agree on + being superior to - for FIP. It's easier to quickly comprehend for whatever reason.

41 minutes ago, Minny505 said:

😂 And this is the conundrum of TV and radio broadcast teams across North America.

That said, I think FIP+ is both contextual and predictive in that it predicts the future performance of a pitcher relative to the league. A FIP+ of 110 predicts that a pitcher will be roughly 10% better than the average pitcher.

Tangent time: One stat I always wish was available readily available is something spFIP+, which was relative to only SPs, and it's counterpart, rpFIP+. For that mater 2BwRC+ for relativity on offense to to positional average. A 90 wRC+ from a catcher is roughly equivalent to a 105 wRC+ for a RF. Either a separate stat or something like the positional average in parenthesis next to the + stat. 

Maybe it's time to dust off the SQL skills...but I wouldn't even know where to get the data sets to begin. 

And FWIW, I agree on + being superior to - for FIP. It's easier to quickly comprehend for whatever reason.

I believe subscription to Stathead from b-ref can get you access to the underlying data.