|
|
Post by goldservice on Oct 4, 2014 9:37:36 GMT
Data: most early enders July-September; A, A+ only; accepted only; 24 loans  Q1 Why do early enders cluster in the second half of the month? (data: 18/24 on or after 14th of month) - Is it related to property auctions clustering in some months, eg Jul/Aug 2014, in the first half of the month? - Perhaps the early closing requests are there but FC dissuades the borrowers? - Or is it that the borrower has a monthly payroll to fund? Q2 What motivation would FC have to talk borrowers out of closing their auctions early? Is it too far fetched to wonder if FC prefers to stop lenders’ money being tied up (in loans that are awaiting acceptance) so that the funds are more likely to trickle into harder-to-fill property loans? Q3 Why do early enders happen on Mon (5/24), Tues (8) or Fri (6) much more often than on Wed (3) or Thu (2)? Q4 Can a borrower execute the closure themselves, or do they have to ask FC to do it? (It looks like the latter as early enders don’t happen at weekends except that in the last six months I once saw an early ender on a Saturday morning, and 8003 ended between Sun 21:00 and Mon 07:55 28-29/9/14 with no bids after the BBB 10% bid of £7814 at 21:05 on Sun 28/9 so I think it ended then - predetermined by borrower? Rejected on 3 Oct.) Q5 Why do some auctions close early and then restart? I’ve seen restarts twice this summer: one restarted after only one hour (13.8%, 1/9/14, loan id n/k), the other after 24 hours; both in office hours. Was the borrower eventually persuaded by FC to restart? Or did it dawn on the borrower that their loan rate could fall further, so they called FC, and … ? Q6 How can you spot a likely early ender? * Perhaps if they have borrowed before (which you can detect easily) and closed early on the previous loan (which you can’t detect easily, I think, unless you’ve already bought the previous loan, but then you may be overexposed if you bid on the new loan). * The size of loan seems to be no predictor. * Perhaps the size of the gap between max bid and average bid is an indicator? Q7 Should early enders over 14% be avoided because it shows desperation? What is their record on early defaults? Q8 Are there any borrowers reading this forum? |
|
fasty
Member of DD Central
Posts: 1,038 
Likes: 388
|
Post by fasty on Oct 4, 2014 10:00:00 GMT
Thanks for the stats, very interesting. I've been giving some thought to the topic of early-enders myself. I managed to cop my first loan parts at 15% by funding a loan just as it was approaching 100%. Since then I've managed to get a few good rates by doing similar. I've been pondering the trade-offs between the potential of higher rates and having additional cash tied up. And as you say, perhaps early-enders are also more desperate. I'm also taking into account the average rate vs. highest rate as it would seem to follow that a borrower who finishes on a crippling average rate is more likely to default.
|
|
ianb
Posts: 161
Likes: 54
|
Post by ianb on Oct 4, 2014 17:21:29 GMT
Interesting indeed. A few thoughts- 1. If you read the borrower manual (which I'd recommend) you see they control the closure. Why they close early ? (a) they are sapps and don't understand the process (b) the think they are better off with the money early than waiting for a lower rate (c) they don't like the Q&A's as they are being discovered as dodgy and see this as a way to get out of the process. For the rare few that end at the weekends, I think (a) is most likely. 2. On timing, I always thought Monday am was best time to pick these up, as gives them most additional time to make use of the money in the week (assuming they need 1/2-1 day to finalise things and get the money). More recently, they do seem more evenly spread, fewer at the end of the week. Time of the month - new one me, maybe to do with month end accounting ? 3. Restarting - hadn't noticed that myself, but would guess (a) and that they've gone back to FC to admit a boo-boo. 4. Spotting them - I hadn't found a way to do so. In the past when I followed FC closer, I thought C's wanting a mid-size loan had a propensity. Definitely those coming back for a second (or more) loan have a tendency to early-end. 5. Are they higher risk - I would guess so but have no stats to back this up, just gut feel. Long may the early ender continue, with my inflated rate of >9% they are all I tend to pick up these days!
|
|
|
|
Post by loanstar on Oct 4, 2014 20:22:04 GMT
Like everyone I have wondered about early enders. Three thoughts.
The borrower may require the money quickly, for good reasons. This is one a the advantage that FC and other P2Ps have over the banks.
In one Q and A a borrower stated that the gross profit on a certain contact would be 33%. So any borrowed money would still
make the contract profitable.
Some loans are being arranged through intermediaries. Obtaining the money at all may be more important that a rate only 1 or 2 % lower.
|
|
|
|
Post by captainconfident on Oct 5, 2014 9:46:52 GMT
Early enders are of immense importance to FC. Without them, there would be an Eb*y situation where there is no point in bidding during the auction. The only bids that mattered would be in a firestorm in the last two minutes, in every loan.
The possibility of early enders is the only reason that anyone bids before the very end. If borrowers wised up and stopped doing it, the platform would probably collapse every time an auction ended.
|
|
|
|
Post by goldservice on Oct 5, 2014 10:25:50 GMT
There may be some bidders who have other commitments and so, even if there were no early enders, they would be unable to bid in the last two minutes though this would only justify bidding up to say 12 hours in advance.
|
|
blender
Member of DD Central
Posts: 5,719
Likes: 4,272
|
Post by blender on Oct 5, 2014 12:54:46 GMT
Yes, it is just conceivable that there may be manual bidders who cannot be present at the end of the auctions that interest them - sad for them. And for the fixed rate property loans, where FC buys the unsold parts, people may bid earlier. Then again there are Autobidders who are bid on behalf of, up front. And there are those who bid early to shut out the Autobidders. Captain Confident overstates it but yes there would be a tendency on the business loans to see where the autobidders take it and to fight it out out at the end.
|
|
|
|
Post by captainconfident on Oct 5, 2014 23:25:29 GMT
I don't think I overstated. I use a snipe program if I want something from EB*y at 12 seconds from the end and sniper software for FC would quickly develop if there were no early enders.
|
|
blender
Member of DD Central
Posts: 5,719
Likes: 4,272
|
Post by blender on Oct 6, 2014 8:39:58 GMT
I don't think I overstated. I use a snipe program if I want something from EB*y at 12 seconds from the end and sniper software for FC would quickly develop if there were no early enders. I was not saying that you had greatly over-stated it CC, just that it is more complex than you suggest. However, I am not so convinced that the development and use of sniping software for FC is being held back by 8 early closers a month at A or A+ (from Goldservice's stats), or about 10% of the marketplace. Though being a manual person who does not bother chasing early closers, I have no personal experience.
|
|
|
|
Post by goldservice on Oct 7, 2014 11:45:23 GMT
I’ve inserted a chart into the OP above.
To collect the data, I bid from the outset on most A and A+ auctions for 3 months. One objective was to see if there was any pattern. There was: 25% of days in the month have 75% of the early enders viz Mon, Tue and Fri in the second half of the month (see chart above). The second objective was to see how valuable the resulting higher rate loans would be and to compare that with the cost of tying up money at 0%. My conclusion was that winning as many early enders as possible might add 1-2% to my overall returns but the amount of capital that I had to tie up was effectively losing me 1-2%. Finally, I think that my sample of only three months was too small (eg see how August is different from July and September in the chart) and so I have an excuse to carry on chasing these beasts with my blunderbuss bidding for another month or two.
|
|
sl125
Member of DD Central
Posts: 85
Likes: 64
|
Post by sl125 on Oct 7, 2014 12:15:50 GMT
Based on your sample, there is no statistical significance in the days of the week that these early enders occur. This is because the sample size is so small, that random clustering will overshadow any statistical significance.
For example, if a rolled a dice 24 times, I may get the following distribution:
1: 3 times
2: 2 times
3: 8 times
4: 4 times
5: 0 times
6: 7 times
Would we conclude that the dice is biased towards 3 and 6, and against 5? If I rolled it a further 24 times and received the following distribution:
1: 2 times
2: 4 times
3: 2 times
4: 4 times
5: 4 times
6: 8 times
What would I conclude?
|
|
ianb
Posts: 161
Likes: 54
|
Post by ianb on Oct 7, 2014 12:36:37 GMT
If you are continuing this exercise I would suggest going beyond A,A+. When I was more in to FC I noticed that more of the higher risk loans ended early, maybe because the low risk ones aren't gagging for the cash and will hang on to the end.
|
|
|
|
Post by captainconfident on Oct 7, 2014 14:08:04 GMT
If you are continuing this exercise I would suggest going beyond A,A+. When I was more in to FC I noticed that more of the higher risk loans ended early, maybe because the low risk ones aren't gagging for the cash and will hang on to the end. Yes, seems so to me too. Also, I think there is a higher possibility of an end to each auction when the headline rate reaches a whole number, 10%, 9% etc, where the borrower has previously decided that this is the rate they would accept. That's my tip for all you big game hunters.
|
|
blender
Member of DD Central
Posts: 5,719
Likes: 4,272
|
Post by blender on Oct 7, 2014 15:46:31 GMT
Hunting higher risk borrowers desperate for cash? A good plan if you can sell them at a high premium before the second payment is due, but I would not touch a C- at 15%.
|
|
ianb
Posts: 161
Likes: 54
|
Post by ianb on Oct 7, 2014 17:53:06 GMT
All this talk about the early enders has them cowering in the bushes I think, I haven't snagged one so far this week.
|
|
