Community Positional Prospect #29
by auclairkeithbc on Dec 11, 2011 12:38 PM EST
Email
PrintWith 23.6% of the open vote and 59.6% of the runoff, George Springer is elected Positional Prospect #28.
*
RESULTS:
George Springer: 23.6% (59.6% In Runoff)
Josh Bell: 23.6% (40.4% In Runoff)
Gary Sanchez: 14.5%
Wilin Rosario: 9.1%
Ryan Lavarnway: 9.1% (write in)
Michael Choice: 7.3%
Starling Marte: 5.5%
Leonys Martin: 1.8%
Gary Brown: 1.8%
Nick Castellanos: 1.8%
Jean Segura: 1.8% (write in)
Tim Wheeler: 0%
*
CANDIDATES: Josh Bell, Gary Sanchez, Wilin Rosario, Ryan Lavarnway, Michael Choice, Starling Marte, Mikie Mahtook, Jonathan Schoop, Jean Segura, Billy Hamilton
*
IN ROTATION: Jedd Gyorko (#27-1.5%), Kolten Wong (#27-1.5%), Will Middlebrooks (#27-1.5%), Leonys Martin (#28-1.8%), Gary Brown (#28-1.8%), Nick Castellanos (#28-1.8%), Tim Wheeler (#28-0%)
*
TESTERS: Rymer Liriano, Eddie Rosario, Mason Williams, Garin Cecchini, Javier Baez, Cory Spangenberg, James Darnell, Marcell Ozuna, Andrelton Simmons, Oswaldo Arcia, Matt Szczur, Aaron Hicks, Joe Panik, Matt Davidson, Derek Norris, Christian Bethancourt, Brandon Jacobs, Robbie Grossman
*
#01 - BRYCE HARPER - 59.2%
#02 - MIKE TROUT - 38.8% (In Poll #1)
#03 - JURICKSON PROFAR - 52.9%
#04 - MANNY MACHADO - 31.5% (55.1% In Runoff)
#05 - DEVIN MESORACO - 45.1%
#06 - WILL MYERS - 31.6% (51.4% In Runoff)
#07 - JESUS MONTERO - 61.3%
#08 - ANTHONY RENDON - 53.1%
#09 - TRAVIS D'ARNAUD - 46.6%
#10 - NOLAN ARENADO - 33.3%
#11 - MIGUEL SANO - 37.3%
#12 - YONDER ALONSO - 22.5% (45.3% In 3-Way Runoff)
#13 - FRANCISCO LINDOR - 26.1%
#14 - BUBBA STARLING - 33.8%
#15 - XANDER BOGAERTS - 25.0% (55.2% In 3-Way Runoff)
#16 - ANTHONY RIZZO - 27.8%
#17 - OSCAR TAVERAS - 40.3%
#18 - BRETT JACKSON - 20.3% (58.7% In Runoff)
#19 - YASMANI GRANDAL - 31.3%
#20 - JAKE MARISNICK - 27.3%
#21 - HAK-JU LEE - 28.0%
#22 - NICK FRANKLIN - 24.3%
#23 - CHESLOR CUTHBERT - 30.3%
#24 - CHRISTIAN YELICH - 27.1%
#25 - JONATHAN SINGLETON - 21.9% (53.8% In Runoff)
#26 - MIKE OLT - 21.5% (67.7% In Runoff)
#27 - ANTHONY GOSE -22.7% (70.5% In Runoff)
#28 - GEORGE SPRINGER - 23.6% (59.6% In Runoff)
*
131 comments
|
1 recs |
Do you like this story?
Comments
+1
"We did a lot of good things last year, and now we've got Julio ... That does nothing but improve the offense, and we expect to do better. That's our goal, to lead the NFL in everything. Every offensive category." -Roddy White
+1
Read Me At: Twitter/Blog/MLBBonusBaby /Giants Nirvana
by Gobroks on Dec 11, 2011 3:37 PM EST via mobile up reply actions
+1
Yoenis Cespedes
http://www.youtube.com/watch?v=aW9ge8l3jY8
by SteveHoffmanSlowey on Dec 11, 2011 7:16 PM EST up reply actions
+1
I don't always like to tweet, but when I do it's @HouseOfTheBB
And when more than 140 characters are needed, I write at HouseOfTheBluebird.com
OTHERS (Post Player Name And Reply +1 To Officially Vote For Non-Listed Players)
by auclairkeithbc on Dec 11, 2011 12:40 PM EST reply actions
because
people wanted him added first.
by auclairkeithbc on Dec 12, 2011 1:41 PM EST up reply actions
Suggestion
It appears you’re laying each list out so the players are listed in order of their prior finish. I’d suggest you mix ‘em up so that the group doesn’t get conditioned to a built-in expectation of the top guy(s) already having the most support and the ones lower down being fringe candidates. If the list is a little more complex to work through, people will reevaluate the choices more.
by siddfynch on Dec 11, 2011 12:53 PM EST reply actions 1 recs
Well
I think there is an element of gathering support that is a good thing in these polls. If it is true that there is an advantage to be listed near the top, then I think the fairest thing to do is list the people who have had the most support in the previous polls. I’d bet there is a small element of truth to what you are suggesting, but randomly lowering one of the popular players while another popular player randomly stays atop the poll, only hurts the integrity of the vote, again, assuming what you suggest is actually happening.
by auclairkeithbc on Dec 11, 2011 1:08 PM EST up reply actions
"small element of truth"
Questionaires are ALWAYS influenced by things like wording and sequence. I’m not suggesting you randomly lower or elevate one guy, I’m saying that if you want people to consider the choices as carefully as you can get, you mix the players up, not conform to a consistent pattern.
There’s no small element of truth here – it’s a very fundamental aspect of human dimensions in research.
Also
Regarding “integrity of the vote”, the vote would have substantially more integrity if people had more tendency to fully scan and consider the range of choices than becoming subliminally accustomed to knowing where to look for guy who everyone else thought was best last time.
you’re doing a great job with these, as i’ve pointed out before. i don’t have a huge concern about the repeat sequence pattern here, but as a research design and stats guy for the last 14 yrs, i can’t not push back on the “small element” and “vote integrity” comments.
Bottom line is that you’ll improve an already good process if the candidates are mixed. For certain.
Eh
It is quite possible that the order has a significant influence on voting. But that isn’t a bad thing, AS LONG AS we keep the ones who have earned the spot in those favorable spots. This system is certainly far from perfect, but it works well because people can lobby for particular players and slowly pick up new voters, gaining momentum along the way. We have a preliminary poll before each poll, which determines the order of the new poll. Then we have the real poll. It isn’t perfect, but players pop up out of nowhere to find themselves ranked #2 or #3 in the new polls all the time. That gives those surging players a nice advantage, and one they deserve in my opinion.
by auclairkeithbc on Dec 11, 2011 2:08 PM EST up reply actions
Interesting
What I hear you saying here is that you DON’T want each event to be independent, but instead you WANT it to be influenced by prior ones. That’s great for things like Bayesian predictions, but here it just means you are arguing for bias.
Why not support an approach that makes voters more likely to re-examine the prior #1 vs #10 more than they currently will? Why try to stack the odds more in favor of any one guy, regardless of past poll performance? I can understand you doing these unintentionally, but why do them deliberately? It just doesn’t make sense to me,sorry.
Well each event is not independent.
There is no way in hell we can achieve that or anything close to that. We have tons of information going into each poll. Mixing up the order, often putting one popular player high on the list and another popular player lower on the list, is not particularly helpful in my opinion. If people really do tend to vote for the players listed near the top, then we’ll just be making it way more likely that one of those two popular players will win and the other will not.
by auclairkeithbc on Dec 11, 2011 3:03 PM EST up reply actions
Independence is a good thing
And the issue is not the information going into each poll, it’s the results of prior events. You WANT it to be as independent of prior results as possible. Whether or not that’s unachievable, it’s certainly improvable.
It was a well-intentioned point, but for whatever reason its being met with odd and somewhat illogical resistance. For example, if the effect I mention is true, then the problem you list above will fade with iteration, not remain (or increase). If there is no effect, then the problem you list above won’t even exist.
I’ll let it go. If anyone else on here has some formal stats training, feel free to chime in here.
In the meantime, I’ll end with just a quick Google of “importance of independence in opinion polling” returns this nugget (with my points in highlights):
Wording of questions
It is well established that the wording of the questions, the order in which they are asked *and the number and form of alternative answers offered can influence results of polls. For instance, the public is *_more likely to indicate support for a person who is described by the operator as one of the “leading candidates”_. This support itself overrides subtle bias for one candidate, as does lumping some candidates in an “other” category or vice versa. Thus comparisons between polls often boil down to the wording of the question. On some issues, question wording can result in quite pronounced differences between surveys.8910 This can also, however, be a result of legitimately conflicted feelings or evolving attitudes, rather than a poorly constructed survey.11
A common technique to control for this bias is to rotate the order in which questions are asked.
The link, with source citations, is below.
http://www.answers.com/topic/public-opinion-polling#ixzz1gGDgP0r6
Just want to weigh in with support for auclair here. This ranking process has to be taken as a whole, not as a bunch of independent events. Also, there’s no way to rotate the order of candidates within a single poll, so you’re exchanging one set of advantages for another where the new set of advantages don’t have any reasonable foundation.
Also, I think voters are far more influenced by the actual prior results than by the ordering in subsequent polls.
Oh my lord
Where to begin?
Of course the ranking process has to be taken as a whole. Eliminating question bias doesn’t affect this. The “independence” issue has nothing to do with the ability to consider the list as a whole. And of COURSE you want the events to be independent – otherwise we’d be using the result of a vote to set the next three or five or however many guys on the list.
And your 3rd sentence contradicts your second. If you want the list taken as a whole, then you’d WANT the new set of “disadvantages” so that the bias evens out over time. Otherwise, you consistently bias things in only one direction across the “ranking process as a whole.” This is the basic principle behind randomizing samples – you still generate biases, but they cancel each other out over time.
And your last sentence is one of those “so whats?” So what if Effect A is larger than Effect B? It’s firmly established that Effect B (question order bias) exists – this has been shown for decades, and is one of the fundamental things to worry about in everything from questionaires to the agendas of controversial meetings. If you can reduce bias from one effect, why not do so? Who cares if Effect A is still out there, as long as you’ve improved the process? There is no credible argument that Effect B may not exist. None.
Oh my lord
Where to begin?
Firstly, you can knock of the arrogant and disrespectful tone in what should be a reasoned discussion of a technical issue.
Regarding the question of independence, I don’t think we actually disagree, and no one is suggesting that a given poll should settle multiple rankings. In order for the polls to be completely autonomous, you’d really have to have a different set of voters wholly ignorant of previous results and discussions, which I’m sure you agree isn’t possible or desirable for our purposes.
My third sentence in my previous comment doesn’t contradict my first, and I defy you to demonstrate otherwise. But I think we do come to the heart of our disagreement in your subsequent comments. If I understand you properly, you see the the initial reaching of some threshold as triggering an accumulation of unmerited advantage against other candidates in subsequent polls, as each higher showing generates a higher vote result in part due to the result and ordering of options. In other words, the initial cadre of enthousiasts trigger a bit of a snowball effect through the subsequent ordering methodology. I think this is a factor, but I doubt the bias is significant down the list (I would expect the advantage of ranking #1 on a list vs. #2 is much larger than the advantage of #4 vs. #7, but I am open to correction). I will return to this “downballot snowball” problem later.
My concern is much more that the elected prospect in a given poll be the reflection, so far as is practicable, of his merits rather than other factors, and in most polls, there are really only two or three possibilities for election. I’d be surprised if a candidate who finished lower than 4th in the previous poll has won the subsequent election. Because we can’t randomize the ordering within a given poll (i.e. have different voters see the list differently ordered) we have the choice of giving one option a random advantage or giving one option an earned advantage (from his placement in the previous poll). While the optimum choice would be to grant neither candidate an advantage, opting for the earned advantage will maximize the integrity of every given election (at least before we consider possible “downballot snowball” effects).
Returning to the question of what I’ve described as a possible “downballot snowball”, I think a few things are necessary to mention: firstly, all candidates, once they reach the threshold level for an ordering advantage to kick in, benefit more or less equally. There could, however, be an advantage for candidates who command a broader diversity of views than others, in that they would reach the threshold level prior to other candidates with a stronger consensus of opinion. Randomizing the downballot order doesn’t resolve the problem however. Because candidates usually go from “in rotation” to election in a fairly short number of iterations (I’d guess 5-7), there isn’t enough repetition for randomness to approximate neutrality between the candidates. Some candidates would randomly receive consistently high slots while others would receive consistently low slots. On balance, I think the downballot problem would be exacerbated by randomization.
One factor I haven’t mentioned is that I think our voting pool is is relatively well-informed such that the impact of a ranking advantage will be lower than in most other polling situations (certainly I think the voters here know more about the relative merits of their candidates than does the average group of people being asked their preference in a congressional primary). This is one further reason why I don’t think that the downballot issues are profound.
I do think there is one more potential issue here: can we order the plausible candidates for election in a manner that counteracts any bias that has accumulated through the downballot process? Presumably that the #2 vote-getter on a given poll is more likely to benefited from the “downballot snowball” any more than the #3 vote-getter (I suspect because he’s likely had more iterations since he reached the threshold for benefiting from the ordering methodology). However, I don’t think this is likely to be a big advantage. Now I think prima facie, the #2 candidate is more likely to be the objectively superior prospect than the #3 candidate, though there will be variation. Now, if we were to reverse the 2/3 candidates order for the subsequent poll, which is the more plausible conclusion should the former #3 win the poll? That the ordering advantage in the final poll overwhelmed the better candidate (who was presumably the former #2) or that the downballot advantages the former #2 had a relative advantage from had been cancelled out, allowing the superior candidate to win? I definitely think the former situation is more likely, so I think the current methodology is at least near-optimal.
by gabrielsyme on Dec 11, 2011 10:06 PM EST up reply actions
Regarding Effect A/Effect B
I accept your principle argument here. Certainly we should be concerned with an effect even if there are larger, intractable effects. What I was hinting at, however a crowding-out effect.
Voter knowledge of prior results and the question order both derive from the same source (prior results) and both act in the same direction. In such a case the dominant factor crowds out the smaller factor. Not to zero, but it will reduce the secondary factor’s effect. In other words, I’d contend without voter knowledge of prior results, question order would have a larger impact.
by gabrielsyme on Dec 11, 2011 10:19 PM EST up reply actions
in my opinion you are looking at this in the wrong way.
this list is in great part about building support, not 100% independent trials. we kick a player off the poll if they get less than 4% of the vote. It is quite possible if I kept them on the list, they might win the following poll, but as a write in candidate, they would have a hard time winning. personally, i don’t think the 4-percenters have a very good chance of winning the following poll, even if I placed them at the top of the following poll. i realize it is a possibility, but we have to kick some players off to give more people a chance. People here are for the most part following these polls, not popping in completely randomly. They will vote for their favorite 4-percenter to keep him on the poll, and when he gets enough support to stay on, he’ll already be in a pretty high spot in the poll, since there is a lot of turnover in these polls. So basically anyone that even gets 3 votes, is going to have a high slot. The top of the list is the guys who stayed on it. The bottom of the list is the new guys. It isn’t a problem. However, manipulating the list in a seemingly random pattern could become a problem. People might not like the runner up from the previous poll getting listed low on the next poll, if they believe that is the way people vote here.
by auclairkeithbc on Dec 11, 2011 6:32 PM EST up reply actions
Forget the "independent trials" issue
You’re misinterpreting what that even means. It’s not even necessary to my original point. And the comments immediately above about how guys get on or off is completely irrelevant to the issue of what you do with them once they get on. Nobody is saying your process for that is flawed.
All of this is just about the following:
Me:
1. You are setting up in the questions improperly by not changing the sequence. This is “question sequence bias”, and is proven by lots of research on the subject. Fortunately, it has an easy fix – rotating the questions.
2. Here are some links to reference my claim.
3. As long as you are spending shit-tons of time on the process, I figured you’d want to at least correct an error.
You:
1. There may be a small element of truth to this effect that is well-accepted within the world of statistics and polling science.
2. However, I believe that the recommended solution of changing the sequence of questions is a artificial manipulation, regardless of what all the experts and textbooks say.
3. Plus, I actually think the runners-up on one poll should get an artificial advantage in the next poll. 4. So, I will actually set the poll up in a way that is the equivalent of giving them an extra 5 or 10 or some other unknown number of fake votes next time, and do so in a way that makes this error consistent.
5. But, for some odd reason, I will continue to contend that the known, accepted solution to question sequence bias is actually worse than my way.
Honestly, don’t take my word for it. Just do a web search. Here’s the top link quote that comes up for me when I Google “question sequence bias”: Read it, please.
One of the biggest hurdles you will face when creating online surveys is dealing with order bias. No matter what kind of online surveys you are creating, whether they are educational, concept testing or even for feedback on an event, the order of the questions can lead to a bias. Once this sequencing bias comes into play, the answers from your online survey respondents stop being representative of their true feelings and start following patterns.
got your back
this is one of those times I wish there was private messaging on SBNation.
by gabrielsyme on Dec 11, 2011 10:09 PM EST up reply actions
auclair
I appreciate all the hard work you have put into this, but don’t belittle that hard work with a comment like that to someone with a legitimate concern.
Now, I don’t think this is that big of a deal here, and I know it is in the real world, sidd. One, I think we have a more informed population on this specific issue, than if we started asking the same people’s opinions of policy, movies, food, etc. There is already a self-selection bias at play, and for the better in this case. Secondly, many of us, me included, are going off our own pre-made list, so arrangement of candidates does not matter (as I have often voted in the “Other” section). Even those that do not have a set list, have general idea of their next 5-10 candidates.
I think that the way auclair has set this up is more as a convenience for previous voters. I honestly don’t remember how this was done in past polls. However, if enough community members agreed with sidd, and felt this was an issue and wanted more random listing, than I think it is something that should be implemented. If it is just you sidd, which so far it is, than I think you have made a good and relevant point, but it would probably be time to leave well enough alone.
by cookiedabookie on Dec 11, 2011 11:48 PM EST up reply actions
Id like to chime in
. . I agree with ak here, even though I respect what Sidd is saying and agree with the logic of much of it. I also think part of ak just lol’ing the whole thing is due to the fact that they just both kind of stopped listening to each others points and, reached an impasse.
Sidd, I can not, and wont dispute the existance of sequence bias but, I, as just one voter actually find our sequencing in these polls usefull in the evaluation process.The fact of someone garnering more votes in the previous polls actually causes me to revaluate the players who I didn’t previously like as much. It makes me look again, based on how much support they are getting. I kind of look at the collective of the community as a learning tool on players and, this is coming from someone who is pretty far off of the mainsteam view of a lot of ballplayers.
Again, I am just one guy but, remember that one vote here is much more powerful than in most elections. I could be anywhere from 1 to 2 percent of any given vote here.
" Kenny G has a Christmas album out this year. Hey, happy birthday, Jesus! Hope you like crap! " - Norm MacDonald
i'm not the one belittling people here
my “lol” response was to a ridiculously insulting post. it was such an absurd attempt to make me look like a fool, that the only response was the one i had.
by auclairkeithbc on Dec 12, 2011 9:20 AM EST up reply actions
FWIW
this kinda stuff does fascinate me. How the human mind works, etc.
by McCutchenIsTheTruth on Dec 11, 2011 10:21 PM EST up reply actions
I'll
come across as a homer, but I think it’s high time Starling Marte gets on this list. Similar profile as Gose, but has had a much higher average, and better K rate at the expense of Gose having a superior walk rate.
They are two years different in age, but my gut just think that’s being overstated and overvalued in Gose’s favor. I’m not saying Marte is the better prospect, but I would group them right beside each other, and it looks like Marte will fall off the pace from Gose.
by McCutchenIsTheTruth on Dec 11, 2011 2:21 PM EST reply actions
I agree
I like Marte and he should get on before Bell for sure. Maybe he will.
by auclairkeithbc on Dec 11, 2011 3:04 PM EST up reply actions
No
he doesn’t. He doesn’t walk, but he doesn’t strike out an alarming rate.
As far as that goes, he’s some what like Robinson Cano
by McCutchenIsTheTruth on Dec 11, 2011 8:08 PM EST up reply actions
You know, Pizza cutter did a stat called sensitivity and response bias
that measures plate discipline. I found the data for the MLB. I’ll try to find it for the minors with minorleaguecentral.com’s plate discipline stats when I have the time.
Aight
I’d like to see that. Sounds interesting. For the record, I’m not saying Marte is without any plate discipline issues, just that I think “awful” is a strong word.
by McCutchenIsTheTruth on Dec 12, 2011 5:47 AM EST up reply actions
I am going to count on your support when I nominate Vitters :)
he walks slightly more and strikes out less, ISOs are pretty much the same, and Vitters is a year younger! (I’m serious about putting Vitters on as a tester around 40-45, by the way.)
by PrincetonCubs on Dec 12, 2011 12:32 PM EST up reply actions
I haven't done those stats yet
but just looking at some plate discipline stats. Marte swings at 25% of the pitches he sees out of the zone, the average batter swings at 19% of the pitches they see out of the zone. That’s pretty awful for Marte. I’ve done some research that swinging strikeouts are worse than looking strikeouts. Marte’s K swinging to K looking ratio is 4.17, the league average is 2.81. Marte swinging at so many pitches out of the zone looks even worse because he dosen’t make contact on those swings. Marte’s Outside of the zone contact% is 55.1%, the league average is 63.2%. Marte’s outside zone swing and miss rate is 11.2%. The league average is 7.14%. Marte does have his strengths. He has very good bat control as on pitches in the zone, Marte’s contact% is 78%, the league average is 67.1% and his overall contact rate beats the average 74% to 66%, but Marte does have awful plate discipline.
I have Gose at 33, Marte at 37
But I also have plenty of other players to support before Marte.
by cookiedabookie on Dec 11, 2011 11:48 PM EST up reply actions
For
positional players or overall?
by McCutchenIsTheTruth on Dec 12, 2011 5:48 AM EST up reply actions
I really don't know why Middlebrooks is still in rotation
Most Red Sox lists, he’s ahead of Bogaerts who we put #30. Most lists he’s around number 40 overall, yet he dosen’t even come close on this list.
Disagree
Older prospect, moderately worrisome K and BB rates, good defensive reports but not without caveats, doesn’t add value on the basepaths. His only really impressive campaign was in Portland this past year, and he then had a horrible month at AAA; so he’s been above-average as a hitter for three straight years, but not really impressive in any one. Not sure why he should be in the conversation yet.
Walking it back...
My last sentence there is, looking at it again, a bit of an overstatement. He should obviously be in the conversation (otherwise why should I bother putting in my more skeptical perspective), but I have difficulty seeing him as better than the current leading candidates.
I would take him over Wilin Rosario
and Gary Sanchez., If you’re complaining about Middlebrooks’ BB and K rates, those guys have absolutely awful BB and K rates. Middlebrooks hit better than Rosario and has a higher ceiling and is closer to the majors than Sanchez.
Well, Wilin has pretty great defense as a catcher
That probably gives him a higher ceiling than Middlebrooks. I could see the argument against Sanchez positionally, if you don’t think he can stick a C, but he has some serious thunder in his bat, one of the best power tools in the minors. So both of them are ahead of Will for me.
by cookiedabookie on Dec 11, 2011 11:52 PM EST up reply actions
Agree. If you’re pessimistic on Sanchez sticking at C, I think Middlebrooks may move ahead of him (and a bunch of other guys)
Yeah, but Sanchez has a much higher offensive ceiling
So it just depends on what you value more, defense or offense, from a prospect.
by cookiedabookie on Dec 12, 2011 12:22 PM EST up reply actions
It seems these lists are much more ceiling heavy
Honestly, Bogaerts shouldn’t be as high as he is, he’s a fantastic prospect but there’s a more than decent shot he never sees meaningful time in the majors due to his age. He could also be a star but there’s just a lot of volatility given his age.
Middlebrooks on the other hand will almost certainly have a major league career. The issue is that his ceiling seems to be a guy who at his peak posts a .250/.320/.500 line with above average defense. While that’s a good player, it’s far from a star.
I tend to believe we prospect watchers sometimes get a little too wrapped up in ceiling.
"a .250/.320/.500 line with above average defense"
Is a solid 4+ WAR player, All Star. That is a star. I do not project Middlebrooks to have that type of power in the majors. I see his ceiling more as an Aaron Boone bat with better defense. A 2.5-3 WAR type player.
by cookiedabookie on Dec 12, 2011 12:21 PM EST up reply actions
I'm talking his potential peak value
and ceiling at that. It’s more than likely he ends up as a 2-3 win player, for sure.
He’s a good prospect, and will almost certainly have a major league career but even if everything goes right for him, he’s probably at best a healthy season Joe Crede during his best seasons.
I suspect a lot of players with higher ceilings outrank him and never sniff the majors… it’s just the nature of the prospecting.
meaning
we as fans tend to underrate players who will likely be useful major league starters on lists like this in favor of the huge talent huge uncertainty guys.
Voting closed
Bell wins with double the votes of the next highest vote-getter.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Minor League Ball
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
