My Criteria

There are very few specific criteria that I expect to see out of a potential admin candidate. Most of my criteria are abstracts:

1) I want to see meaningful edits. (e.g. using twinkle/huggle/bots don't count. 20,000 edits in 3 months does not impress me, in fact it may be grounds for an oppose.)
- While a large number of edits doesn't impress me, too few might discourage me... I generally want to see 3000-3500 MANUAL edits (EG not huggle, twinkle, or the bot of the day.)
2) I want to see a user who has been active for 5 of the last 6 months. (I define active as having about 150 edits in a month.)
3) I want to see a user who has shown an understanding of policy and guidelines. The most common way to do this is through XfDs but this is not the only way.
4) I want to see a user who has helped build the encyclopedia. The most common way to do this is through article creation. Again, that is not the only way.
5) I want to see a user who is civil. Incivility is the quickest way to garner an oppose.
6) I want to see consensus-building in some form. Communication between the editor and others is perhaps the most important thing I want to see! I will pretty much ignore any and every other consideration for a person who is sought out by fellow wikipedians for their help/guidance. Being an admin is not about blocking/protecting/using the tools, it is a sign of trust and cluefulness. A person who is sought out by the community is trusted and clueful!

Other factors:

A) Has the user been blocked in the past 12 months? Blocks may or may not have any actual impact on an !vote. If the candidate had been blocked in the past year, I will explore why. Was the block justified? How has the candidate changed since then? How did the candidate respond to the block? etc. As I believe the review period for candidates can be up to a year, I'll look for that. Anything older than a year, I will ignore--even vandalism blocks!
B) Does the user take part in an unusual area of wikipedia? (I love supporting people who fill niches that nobody else cares for!)
C) Has the user undergone a previous RfA? If so, did they address the concerns of the first RfA? (I am willing to overlook some flaws in a candidate if they can show me that they listened to and learned from their previous RfA.) Also, did the user wait 3 months before running again?
D) Who nominated you? A good nomination does matter! People are more critical of the self-nom or nom by unknown than they are of a recognized admin. If a strong admin nominates you, you get to ride that admin's reputation and the belief that the admin properly vetted you. An unknown nominator raises questions. Likewise, while the RfA is supposed to be about the candidate, a nom by somebody disruptive (even an admin) will possibly count against you... at bare minimum, it will garner closer scrutiny.
E) Did the candidate answer the questions well? This is an interview, give strong answers and we'll probably spend less time researching your contribs! Remember we are looking for a reason to accept or reject you. Give us one early on by giving good answers.
F) Have you prepared for adminship? The most noticeable way is via coaching, but that is not the only way. Some people may, on their own, explore other areas to get exposure prior to an RfA.
G) Do you act like an admin---the tools are nice, but they aren't what being an admin is all about. People who 'act like an admin' are likely to get a support.
H) I do not hold anything against a candidate that was over 12 months ago when running---assuming the issues from that long ago are not ongoing/indicative of a continued pattern. I believe in redemption---and I would rather be able to identify the reformed POV pusher/vandal, than encourage them to "start over" with unknown identities. Similarly, you may have been the best contributor to Wikipedia a year ago, but if you haven't been active lately, then I'm not going to give you credit. The time period can be shorter than a year, if the candidate shows commitment and solid contributions since the indescretion, the year is the maximum that I review. Generally, I don't even review that far back.
I) If you are a speedy deleter, you need to be aware that I hate speedy deleters and will only nominate/support those that I think are among the best. Speedy Deletion is a necessary evil of wikipedia, but you have to be near perfect to get my nom. I feel this way because a single careless speedy deleter can do more irreparable harm to wikipedia than the worst ten vandals.
J) Name changes. If a person undergoes a name change in the past 3 months, I will investigate said name change. I've noticed that some people will change their name to hide from people they've had issues with.---Balloonman ^PoppaBalloonCSD Survey Results 19:21, 13 January 2009 (UTC)

How I Review Candidates

Unless the candidate is clearly not qualified, I will generally spend 1-4 hours reviewing a potential candidate before !voting on the candidate. If I am going to nom them, it will be 2-6 hours! Reviewing a candidate is more of an art, than a science, but I will generally approach my review along the following lines:

I will generally start by pulling up their edit history via this tool. I will check out the total number of edits and what 'space' their edits occur in. I will check to see if they've worked on a few articles seriously or if they are the type that makes only a few edits here and there. I pay particularly close attention to what Wikispace they've worked in and their contributions to talk pages. Being involved in talk page discussions is a strong indicator of how well the candidate works with others. Assuming no problems at this stage:
I will then check their User Page. This is usually a quick review, but I want to know who I am dealing with and what their interest are. Is there anything (such as an offensive user box or message) that is a red flag? Have they contributed to article building/FA/GA/DYK? Does it have a "MySpace/Facebook" feel? Assuming no problems at this stage:
I will then check out their talk page and archives for the past year. Are they civil? Do people treat them like admins? Do people seek them out? Are they being asked questions or are others coming to the page to tell them their mistakes? How do they respond to criticism? Do they have any warning templates? How about notices of speedy deletion or licensing issues? What can I learn about the candidate from their talk page? Hopefully, they respond to the comments on their talk page and not the other persons (responding on other people's pages can leave an incorrect impression with the reviewer and will generally add an hour or more to the review time.) (NOTE: I only care about the past years worth of edits. If you blew it more than a year ago, I'll forgive you, likewise if all of your great contributions were over a year ago, they don't count for you.) Assuming no problems at this stage:
I will check their contributions. Do they use a tool such as Twinkle/Huggle or are most of their edits manual? Use of a tool doesn't disqualify a candidate, it just means that I have to look harder to find their meaningful edits.
I will then look at their involvement on the various talk pages and wikispace talk pages. Again, how do they deal with discussions? What are their contributions? How do they handle disputes/conflicts? Assuming no problems at this stage:
I will then look at their contributions to the wikispace. Are they constructive? What involvement have they had in XfDs? Other adminly areas? Assuming no problems at this stage:
Finally, I will review their article building. I usually don't spend much time here as I should already have a solid understanding of how others, more familiar with them, perceive their work from their participation in various talk forums. Assuming no problems at this stage, I will support or oppose as appropriate.

RfA Questions

While the basic 3 questions are important, I generally do not pay too much attention to generic questions asked of the candidate. If there is a specific question asked about a specific issue for a candidate, then I will read those questions with interest. Similarly, the only time I ask questions of a candidate is if I am truly sitting on the fence and want the candidate to convince me to support them. Sometimes the best way to do this is to admit fault or that an error was made.

How I !vote

Strong Support I rarely give out strong support. Strong Support means that I am familiar with the candidate and have spent a lot of time reviewing them. Generally Strong Support is reserved for admin coachees---but that is more because I've spent scores of hours reviewing their edits. A strong support means that I've known the candidate for months AND have given them a complete review and not found any problems. If you are a candidate I first learned about via RfA, then you won't get a strong support. Strong support is reserved for people whom I have grown to trust over time and had that trust reaffirmed via a detailed review.

Support In order for you to get a support it means that I have spent 1-4 hours reviewing your edits in detail and didn't find anything that I couldn't live with.

Weak Support usually means that I was too lazy to give a complete review of the candidate. If that is the case, I will indicate that is the reason why. I give weak support when nothing jumps out at me, but I've probably only spent an 20-30 minutes reviewing them. I use weak support because I don't want others to rely upon my review. It could also mean that I am overall pleased, but that there were some concerns.

Neutrals are fairly rare. They are usually an indicator that there were both plusses and minuses to the candidate. It is also a strong indicator that I WANTED to support and saw something that I didn't like. Or a candidate where reason said I should oppose, but something made me want to support. It is usually used when logic and emotion are at odds over a specific candidate.

Weak Oppose usually means that you don't have the experience that I am looking for (length of tenure and/or number of edits) but that you are otherwise a solid candidate. Weak oppose may also be based upon a gut or bad feeling. Generally, I won't spend too much time justifying a weak oppose---I figure it's a weak oppose for a reason.

Oppose usually means that you are failing in some way. I will generally write out a detailed explanation as to why I give an oppose. I will often spend more time on a candidate than when I support.

Strong oppose I only give this where I honestly believe that passing the candidate would be a mistake. I will never give a strong oppose without giving firm reasons for the !vote and diffs/examples supporting the position. I will generally spend the most time reviewing candidates with a strong oppose as I want to ensure that it is appropriate.

As I spend so much time (1-6 hours) reviewing candidates, I don't always vote on every candidate. The only time that I generally !vote on candidates is if I am one of the first 10-20 !voters. The only exception is if the candidate has a close/contentious RfA. In other words, if the candidate is at 40/1/0 I probably will not review the candidate. If, however, the candidate is at 40/15/10 I probably will. I do not !support to pile on supports. If I support, it is because I reviewed the candidate myself and feel comfortable doing so (weak supports may be an exception). Thus, I rarely !vote where the outcome appears to be set one way or another---it's not worth the time. Because of this, I probably oppose a larger percentage of the time than I support---most of the candidates who are qualified will pass without my support. I tend to partake in candidates who are less well known.

Nominees

When I nominate candidates, I will not !vote until the last day of their RfA (or if they reach the 100 milestone). I do this because nominator support goes without saying and I always think that it is silly to take pride in beating the nom. Beat the Nom support has traditionally meant that the candidate was so strong that they beat the nom. As I want my candidates to be as strong as they can, I take my time so that EVERYBODY beats the nom!

Ok, in all seriousness, I take my time because I think people who take pride in beating the nom are often short-changing their investigation into the candidate and I really do think "beat the nom support" is kind of silly. It no longer means that the candidate is so strong that you can support on name recognition alone---for many it is simply a goal to "beat the nom." Yes, this is a littly pointy, but in a non-destructive manner.

Plus, it shows faith that you believe your candidate will pass without having to stack the deck from the getgo.