Random rankings thoughts

Since we started the THT rankings process, I’ve learned two things. Rankings don’t mean anything because they don’t take into context roster formation, inflation/deflation, or any other variables that render themselves during the draft process. You should only use rankings as a frame of reference. I wouldn’t even recommend using them as a base for drafting because everything changes at the draft table, even down to a silly list.

For example, I wouldn’t draft Johan Santana based on my rankings, but if my team conditions favor an ERA upside play, then there may not be a better, cheaper option than Santana. I obviously would pull the trigger no matter my previous feelings or, more importantly, my rankings.

The other thing I learned is that I am fascinated by standard deviation in “expert” rankings. I will spend the most time on this topic because I think it has a real use for you as well as my mindless musings.

Basically, for all you non-mathletes out there, standard deviation is the movement from the mean (average). Thanks to Fantasy Pros , we can actually look at how the greatest minds in the industry feel about a player, even if it’s in an arbitrary ranking. I firmly believe the standard deviation table may be of more use than the overall rankings list itself.

One of the biggest question marks I have when I go into a draft is how other managers truly value players. Obviously, I feel pretty confident about the players I want and how I’m going to go about getting them. The unknown is how the other managers feel about those same players.

By looking at the standard deviation, we are able to understand draft volatility a little bit better than we may have ever before. If a player has an extreme movement from the “expert” consensus thought, then there’s a chance that somebody in your league may share the positive/negative feelings surrounding the player.

Understanding a player with extreme volatility could be a huge bonus for a manager that knows how to use this information properly. I like using a volatile player early in the nomination process of an auction draft. The idea is that the draft room will either overspend or underspend. Either way, that benefits you. People, this is good stuff. I hope you are paying attention.

Here is what I’m going to call the “Kipnis Effect”. Another benefit of using standard deviation as a draft tool may be the way it can enhance your strategy towards certain players that you feel favorably about. For example, take a look at Jason Kipnis. He ranges from 62 overall all the way to 250 with a deviation from the mean of 50.4.

Looking a little further into who has him ranked where, Kipnis becomes an even more interesting case. His ranking amongst individual experts is down at 163 at its lowest. The 174 and 250 rankings are that of CBS Sports and ESPN’s site rankings, respectively. I take this to mean that Kipnis will inherently be cheaper on these sites.

There have been studies, I believe performed by Rotoauthority or Baseball HQ, that confirm draft prices are dictated by site rankings. So if you like Kipnis at 62 overall, you won’t have to spend nearly that much to get him. This is all assuming you aren’t drafting with Kipnis himself or someone related to Kipnis or someone that’s crushing on Kipnis. Here are some other players I found experiencing the Kipnis Effect: Alejandro De Aza, Seth Smith, and J.P. Arencibia.

Just like extreme deviation is important, I think no deviation at all has equal value. Assuming multiple sources provide the same rating, you can have a better idea of what the room is willing to pay for that particular player. I don’t think there would be a more valuable tool than to know nearly every move my opponents are going to make before they make it.

It’s not as easy to garner useful information off non-deviating rankings since most of the pertaining players are the earlier-round selections where consensus thought tends to reign supreme, or they are found in the last three hundred where the expert rankings become fewer. If you focus on the guys in the forty-and-beyond range, you might find a little more useful consistency.

For example, David Price has the lowest deviation of 8.5 with a consensus ranking of 47. I think that you can expect Price to stay in this general area of drafts. For me, a guy that likes Price as a fantasy ace this year, this information is priceless. I can safely assume I will be able to draft him as the 40th player and still find value without much risk of losing him to other owners.

Now, this practice may be reserved more towards snake drafts, but the principle can be used in auction, as well, if you focus your attention on the positional rankings.

Put a pen to paper and map out the players you want on your fantasy team. Winning fantasy baseball isn’t necessarily how much you know, it’s how much work you are willing to put into winning. I believe there’s less pre-work that needs to be done in a snake draft than an auction draft. So, using standard deviation in these postional rankings should help with designing tiers more efficiently in auction leagues. Effective use of tiers is all up to a particular manager’s roster-building strategy.

I actually learned through this exercise with rankings deviation that there is a strong chance I will draft Carlos Gonzalez. Upon looking at where I have him ranked in contrast with my colleagues, I apparently like him more, and he will baseline a tier I fully expect to draft early in all leagues, including my snake and auction leagues.

I must close with my final thoughts about the whole THT rankings process. I always loved my own personal rankings as a basis for thought and have never had to be held accountable for their accuracy. So this whole exercise was new and different. However, I didn’t realize how difficult it would be.

I spend an average of two hours a day totally dedicated to reading, analyzing, and researching fantasy baseball. For these rankings I spent at least six to eight hours of concentrated thinking. Even with that effort, I don’t think they are anywhere near where they should be. I pride myself on the knowledge base I’ve built, which is why I began writing for THT in the first place.

The ranking of players made me realize that any one person’s personal rankings are fatally flawed. We are influenced by the baseline rankings provided by Fantasy Pros. We are slaves to our own bias. I found myself driven to the point of frustration over whether Colby Rasmus should be higher than Brandon Belt when it really doesn’t matter.

Furthermore, I know that my fellow writers would agree that rankings do not and will never account for shifts in draft momentum, nor do they adapt themselves to the fluidity of your lineup. Rankings are cold and dead. Tiers are better, but I don’t think that being constrained by making sure a specific tier is targeted is that much better, though I like the idea of building talent tiers. I figure the name of the game in fantasy baseball is all about accruing stats in the most efficient manner.

Obviously, the most efficient manner to accrue stats is to draft the perfect team and never have to manage or do anything. Now, that wouldn’t be fun, but I started thinking about all the talent monikers we bestow on different types of fantasy players. Now, this is just in the early stages of my thought process, but shouldn’t it be smarter, especially in deeper leagues, to draft based on a specific type of talent rather than a blank name ranking?

Take the famed “five-tool” fantasy player. You could easily tier that out with Matt Kemp, Ryan Braun, Carlos Gonzalez, Justin Upton, and Jacoby Ellsbury in the first tier. Follow that up with an Ian Kinsler, Hanley Ramirez, Curtis Granderson, David Wright, and Andrew McCutchen tier. Then you have the likes of Brett Lawrie, Hunter Pence, and such. You get the idea.

This exercise can be repeated with power, speed, average/contact, RBI (batting order placement), prospects, wins (good teams), strikeouts, ERA/WHIP, and even five-tool pitchers (do-everything talents). I believe that I have always done this type of rationalization in drafts before now, but it was always done in my head as I was making quick decisions. All this list-making made me really feel like some lists are more beneficial than others.

Do these kinds of rankings rather than the ones we did. The Baseball Forecaster by Ron Shandler does a decent job of presenting data in previous stats-based tables that mimic the kind of tiers I’m talking about. However, you have to build these talent tiers yourself.

No one can account for the future, and no one can account for the daily changes in fantasy. Nailing down where to put which player in which talent tier can only come with research. I hope over the month of March to focus on building some of these talent tiers for you guys. Until then, happy spring training to all.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Let there be news – Volume 10
Next: This week in (fantasy) baseball 2/20-2/26 »


  1. SCIENCE! said...

    This made me wonder… how much does your first pick influence the rest of your draft options. And is there a trend here? Could you try to predict your opponents subsequent picks as early as the first round?

  2. Ben Pritchett said...

    @bohman- Thanks, I had an epiphany that in order to best accrue a specific stat. You need to know the guys that get that particular stat well enough that you aren’t scrambling in the draft to come up with the name.

  3. Ben Pritchett said...

    @SCIENCE!- It’s a good thought. The first pick should have a direct influence on how you draft the rest of the way. It’s pivotal to determining you stat-accruing strategy.

    Now, it would be rather lofty to assume you could predict your opponents picks. The closest thing we have is these rankings/standard deviations and average draft positions. Knowing someone’s style would take an incredible amount of time. The theory of narrowing down where a player will be drafted or how much he’ll cost is very possible. You’ll never be perfect so it could only be a tool not a crutch or strategy.

  4. Pochucker said...

    Won’t plug another site but let me just say on another site one of the regular posters provides a great tool.
      He compiles ADP from many many sites so individual site rankings dont over influence final ADP like you get if you just look at Yahoo adp or MDC’s adp. He calls it MDP (median draft position) and updates it regularly once all services start drafting.
      It services as great tool to see trends and is my check off list. Used right (like any adp list) can be valuable. In fact I print it out and distribute to my whole league (now I can really use it to my advantage).

  5. Derek Ambrosino said...

    One strategy I’ve been prone to use in auctions is to nominate expensive players I don’t really want early. …Get other people to spend their money. Sometimes that makes them reluctant to pull the trigger on mid tier talent because they then start to worry about wrecking their budgets, so mid tier guys start becoming bargains.

    The best case scenario, though it’s hard to force, is to have context and the “Kipnis effect” merge. So, you nominate Kipnis in the middle of a run on players of the same skill set or position (I’m thinking Dee Gordon would work here too). That way, you have the guys who are willing to pay because they’re believers and the guys who are willing to pay because they are scared of being shut out of a good option in a position or category. The thing with the Kipnis effect is that it’s like an ebay auction – the fact that I’m willing to pay way more than the current price is no good to the seller unless there’s somebody else to drive my bid up.

  6. Jfish357 said...

    Nominating expensive talent can also backfire, when Matt kemp goes for 36 and 8 pix later tulo goes for 43. Or when inflation isn’t factored right. I’ve often found bargains at the top of the draft. You just have to adapt to where the value is.  Every keeper league shifts over time if the owners are any good. One year its rookies, then 2 years later it’s undervalued vets. A couple years later it’s age 27 guys again.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>