I’m scratching my head at why one should be forced to subscribe to a premium plan in order to increase the pledge on a goal and potentially give more $ to Beeminder when he/she derails 
It’s like saying “we don’t really want to get money from you and we’ll make your life harder if you try to” 
@dreev ?
The rationale for making short-circuiting premium is that if you can jump straight to an amount that you know you’ll never derail at then Beeminder never gets paid. It’s counterintuitive, I know, but it seems to really work that way in practice. We talk about it more in our original blog post on short-circuiting (note the 3 exceptions to the principle of letting you pledge any amount you want).
But funny you should mention this because just this weekend our daughter, @faire, wanted to short-circuit to $10, hit that very pop-up, and it wasn’t at all clear to her that she was only being denied the immediate short-circuiting, that what she meant to do was move the pledge cap so that on her next derailment the pledge would go up. I think we’re leaving a lot of money on the table and harming the effectiveness of Beeminder with confusion like that.
In fact, let me cc @joshpitzalis here so we can think about whether we can more fundamentally fix that UI confusion in the ongoing redesign. Maybe by making the pledge cap the first obvious thing to adjust when you’re not satisfied with the current pledge. If not then I’ll call this a #uvi idea for myself to at least improve the copy.
Huge thanks to @apolyton and @faire for pointing out the problem here!
PS: And to answer @apolyton’s question, in a quite real sense we don’t want to get more money from you. Not in the short term. What’s in our long-term interest is just making you awesomer. When you’re so much awesomer that the $5 and $10 derailments aren’t motivating enough then, yeah, at that point we’ve earned those pledges! (Beeminder’s model is much less perverse than it seems.)
I just woke up but the two quotes seem to contradict each other 
That said, most people betting high against themselves, fail on their bets.
So allowing someone to increase a 5$ to 50$ , would probably be more beneficial (in terms of cashflow) to Beeminder.
By the way, when you start a goal you (i think) can put whatever amount you want, no matter your premium status.
But if you start low, then you can not increase it but rather you to have fake derails? Sound unproductive.
A good compromise would be the popup asked “you can’t increase the pledge now (we want you to take it slow), but you can increase the pledge cap for next time, click here (one click) to make it happen”
And perhaps you might want to communicate the pledge cap right in the goal page as well. You’ll probably reduce questions like “why did my pledge increase” (from people who don’t realise there is a pledge cap)
@apolyton forget 50 usd. That is peanuts. Would you skip going to the gym if you had put two weeks salaries on the line? A months salary?
Not being able to set your own limit on the contract is a problem, I agree, but Beeminder has to make money somehow so that they can stay in business long enough for all of us to be so awesome that 5 or 10 usd doesn’t matter.
As for the contradiction you mention, I don’t see one, except that Beeminder has a long term interest, which is making everybody able to achieve any goal they set their minds on (an insane superpower, now that I think about it), and a short term interest, which is staying in business, that sometimes don’t entirely overlap.
But hey now that I think about it, Beemium is 32 usd/month and allows you to short-circuit, and 32 usd is not that bad for a super-power, is it?
Disclaimer: I don’t have any direct interest in Beeminder, financial or otherwise, except that I fully believe the idea of practical commitment contracts as a service essential to the survival of the free western world (100% not kidding).
Apart from the problem of ensuring that Beeminder is a sustainable business, I personally agree that short-circuiting should be free. In order for that to be workable, then pledges can’t be the primary source of revenue.
Hence including this ability (a superpower, says @tomjen) in a subscription plan.
In the same vein, not everything that I track using Beeminder even needs to have $5 pledged against it. Again, something that’s bundled in a subscription plan.
More to the point: climbing the pledge schedule takes too long.
If you accept a week’s flat spot after every derailment, then you might have an entire month of wasted awesomeness before the pledges reach a level that would start to significantly impact your life.
Remember that you can always short-circuit by entering fake data to derail yourself. e.g. if you were going to spend $32/month to short-circuit to (say) $270, you can do this by paying $135 in derails. That’s only 4 months of Beemium so it’s likely that just paying the $135 is cheaper unless you’re doing this a lot.
It’s probably faster to email support and just ask us to charge you $X and up your pledge behind the scenes. Paying to skip pledge levels actually used to be a feature, but got axed a while back.
Or (gasp!) you could sign up for Beemium for one month, adjust your pledges, and then drop back down to Plan Bee or whatever suits your functional/operational needs.
If you do Beemium for one month, you could pre-create a bunch of goals at high pledge values for whenever you need them later, except you can’t change their slugs without premium either haha
I (seriously) endorse any such shenanigans if it reduces temptation to fake data. 
Fake data is the absolute worst thing ever in Beeminderland.
We definitely explicitly encourage you to get a month of Beemium and immediately cancel it (you don’t have to wait a month – by paying once you’ve paid for that month and we don’t take it away if you cancel, we just don’t renew it next month) in order to do some one-time short-circuiting of pledges, etc.