[brother7]

Has anyone kept a log of P2P-Picks breakdown by grade and Top X%, similar to the attached XLS worksheet?
I think it would be helpful to determine investment/note when trying to deploy a given amount within a certain time frame.

[lascott]

That is clever but it sure seems to vary a lot and especially in the past couple months with the feeding time changes.

Perhaps I will try a spreadsheet like that tho when I transfer some more money at the beginning of the month. It only makes sense if you have an excess cash account where there are not artificial limits like getting money back in your account from "X Note(s) did not issue" emails we get each morning.

My primary purchases are with BV+P2PPicks but I use LR as well.  It is easier to estimate with LendingRobot as they put out estimates and it is one of the coolest features I like. So I tweak my rules criteria (changing "dials" based on grade and term) to get estimated numbers I want.


.

[TravelingPennies]

from: brother7 on August 27, 2014, 04:17:16 AM

[Rob L]

P2PP began providing API access to its picks in late October last year. Before that it was all web browser based.
You had to "buy" a pick to learn the actual LC loan ID. The P2PP API provides LC loan ID's directly.

So, the best one could hope to do, if one accessed picks via the API and logged them each LC release, would be to go back about 9 months.

[brycemason]

Or I could simply run the algorithm on historical originations (not applications--too cumbersome) and provide some aggregate numbers. The algorithm has changed a number of times over the prior year. The most-recent iteration might be best.

[TravelingPennies]

That would be great, Mr Mason!
Instead of a snapshot in time, if you were to post weekly or monthly numbers on an ongoing basis on your website, I think others would find this very useful to gauge their future investment/note.

Just curious... do you keep a historical record of which loans appear on the P2P-Picks API? If you did, that would provide the complete list of P2P-Picks applications (which would include loans that fall out of funding), correct?

[TravelingPennies]

Yes, it's in a database. It's much easier just to apply the algo on the quarterly origination data.

[TravelingPennies]

Decided to code a Python program to make my life easier.