I know the professors are a busy tribe. I would like to save them some time by writing a letter of recommendation myself and sending them for approval.
Would that be seen as impolite as I am evaluating myself and therefore it is biased? It could be regarded as me trying to cheat.
I would like to suggest you ask this professor, if they would be willing to write a strong recommendation for you. Then you would know from their response as to their preference. If they are willing to do so, you may ask if they need anything, your transcript, CV… etc. and if they want you to draft one yourself, they will let you know.
That said, I remember at one time in certain Asian countries, it was usual to do exactly as you suggested. But in the United States, I don’t think that’s the norm.
Ideally, you want a faculty who knows you and your work well enough, such that writing a recommendation letter with evidence of support or vouch for your qualification and potential should take very little time.
Professors do not form a homogenous group. So it is hard to tell if the ones you ask to write a letter of recommendation for you will perceive a draft that you send them as impolite. I know some who regularly ask for detailled input for their letters because they think they cannot manage otherwise to write a timely letter. Others strictly do not want any input and would not use any that they might get. Anything between these extremes exists.
I think there would generally be no harm in asking if you can support in any way, and then do as the letter writers suggest. This also saves you from wasting time to prepare drafts that are then just ignored.
As a side note: The letters I have seen that were written by the recommenders themselves were usually much stronger and more convincing. If a professor wants to write a letter based heavily on your input, it might be wise to look for alternative letter writers.
When asking a person for a letter of recommendation then also ask if you should send them a "draft" for the letter as well. This draft should be in an editable format so that the person can adjust sections he/she doesn't like easily - this is how most people I know do it.
I think there is some danger that a professor might be offended if you offered the letter unasked for. But, you could, offer to provide a draft of a letter, or a set of bullet points, if that would be helpful.
I'd guess most people prefer to write their own letters and some institutions "forbid" candidates to write their own recommendations, though there is no way to prevent it.
But I wouldn't assume that the professor would see it as a positive thing, even if they are busy.
I believe providing a draft letter is absolutely fine, provided you ask your recommender if they would welcome that, dropping the idea if they do not, and make it perfectly clear that they can edit it or throw it out, as they see fit.
In fact, only YOU really know what your entire application portfolio looks like, and only YOU really know what part of your portfolio you'd like each letter writer to validate, so drafting a letter actually makes sense.
You're not "sending it to them for approval". They are the letter writers, and they get to do whatever they want to do with your draft. In case there is any miscommunication in your original question, it is NOT OK to send in the recommendation on their behalf, posing as the recommender. That's a very serious falsification.
As explained in other answers, this is not a good idea. However that doesn’t mean you can’t do things to make your letter writers lives easier by giving them well-prepared summaries. To give one example, for tenure and promotion in my department we send the letter writers an “annotated bibliography” with paragraph about each paper. This kind of concise summary of your work can be very helpful for letter writers, just don’t do it in the form of a recommendation letter.
I would personally never sign such a letter and would look disapprovingly on anyone taking such an initiative.
It is part of my job to agree to write reference letters, and if I commit to do so I do this on my own. Sure: I will often discuss multiple aspects of my letter with a student, to make sure there is no misunderstanding between their objectives and what I want to highlight.
However, my name and hence my reputation is literally on the line and this is NOT something I let someone else do on my behalf. Conversely, I would immediately stop trusting recommendations from a colleague should I learn he or she allowed a student to write a LoR on their behalf.
This sounds somewhat unethical to me. I don't think you can literally just send them the letter and say 'I've done the letter for you, just put your name on it'.
You could instead send them a coherent 2 - 3 page summary of your work (past, present, future) along with your CV which will help to strongly guide them in writing the letter of recommendation.
I think it is extremely impolite.
Letters of recommendation are things they write (or dictate) all the time. It is better that they phrase it themselves.
I think it absolutely outrageous that you would ask a professor to sign off such a pre-drafted letter.
If I were made such a proposition I'd be of a mind to refuse.
It is alright if the professor ask you to do so, because the professor will read the recommendation letter before signing it. If there is any exaggeration then the professor will amend the letter before sending it to destiny. Otherwise you should not do it. However, there is a positive side of this method, as well, in the sense that the recommendation letter often depends on the specific project (research project or PhD or academia job etc). So depending on the project, the candidate is much associated with the project, and hence the candidate, often, is in better position to express the letter.
