Author Topic: P2P-Picks is off line  (Read 9491 times)

brycemason

  • Hero Member
  • *****
  • Posts: 801
    • View Profile
    • P2P-Picks.com
    • Email
P2P-Picks is off line
« on: June 30, 2015, 04:31:06 PM »
EOL

Rob L

  • Hero Member
  • *****
  • Posts: 2060
    • View Profile
Re: P2P-Picks is off line
« Reply #1 on: June 30, 2015, 05:34:00 PM »
RIP

Rob L

  • Hero Member
  • *****
  • Posts: 2060
    • View Profile
Re: P2P-Picks is off line
« Reply #2 on: June 30, 2015, 09:04:06 PM »
A user's post mortem to P2P Picks:

Started using P2P Picks buying LC notes in May 2013 furiously pounding my mouse at feeding times. Think I had the initial $100k manually invested through the web interface within 3-4 months. Can you imagine doing that now? Back then there were four loan bins; Top 1%, 5%, 10% and 25%. Relatively late in my initial investment cycle (fortunately) I had the bright idea to scale note size to loan bin ($100 to 1%, $75 to 5%, $50 to 10% and $25 to 25%). That did not work out very well. There were not enough 1% loans to get much diversification and they were the highest risk (variance); typically Grade F. Bryce must have come to the same conclusion as he combined the 1% and 5% bins, eliminating 1%. It was clear my note size scaling idea was a loser. The D's and E's were performing well so I began investing $25 per note in those in the the top 5% and 10% bins. Later I went to $50 notes, and finally I added $25 notes in the 25% bin to keep the cash invested. As competition heated up for high interest rate notes I was lucky enough to have the the rudimentary programming skills needed to auto buy selections from the P2PP web page and then auto-invest them with LC using their SOAP API. As the arms race accelerated P2P Picks provided an API and LC moved to a REST API; I went with the flow.

Basically I now think there are quite a number of real dog loans out there in every release and then everything else. Avoid the dogs and things will work out just fine. At least that's my hope.
And, oh yes, diversify.

So, for the record, here's one P2P Picks user's results on the closing day of the P2P Picks web site; hope you find it interesting:








Chart by Interest Radar


Chart by Interest Radar
Note the PMAX 1 portfolio didn't (hasn't) worked out well.






edward

  • Full Member
  • ***
  • Posts: 195
    • View Profile
Re: P2P-Picks is off line
« Reply #3 on: July 01, 2015, 02:18:22 PM »
Under your Portfolio names, what does the "NM" stand for in the PMAX selections? And thanks so much for sharing all your information. It's always so informative to see this type info from experienced investors for comparison.

Rob L

  • Hero Member
  • *****
  • Posts: 2060
    • View Profile
Re: P2P-Picks is off line
« Reply #4 on: July 01, 2015, 07:26:48 PM »
According to the comments in my auto-invest code on 7/19/2014 P2P Picks released new models.
I really don't remember the specifics, but Bryce could chime in here with a bit of clarification.
Anyway, at that point I created the NM  portfolios to track the new model performance versus the old.
Don't remember thinking it was much of a big deal at the time, but I'll bet there's something here on LA that will illuminate the point.
For the record the picks ( 1%, 5%, 10% and 25%) were originally ordered from best model score to worst.
The 1% picks were combined with the 5%. Think I remember Bryce commenting on the hazards of reinvesting skewed heavily toward 1%.
Personally I consider F notes unpredictable (in the numbers available) and simply no longer buy any.
I believe Bryce later randomized the picks within each bin (5% 10% and 25%) before posting them on the website (or API).

thezfunk

  • Sr. Member
  • ****
  • Posts: 316
    • View Profile
Re: P2P-Picks is off line
« Reply #5 on: July 02, 2015, 01:04:27 PM »
I don't feel like posting all the screen shots unless there is a huge uproar to do so but I can tell you P2P-Picks had me earning ~17% with primary notes only and then I started selling distressed notes on the secondary market so my combined return was ~12%.  My trading only portion was ~4.5% but that was coming up as I started getting out ahead of distressed notes in that instead of waiting for them to go bad I would sell anything with a FICO score drop of 50 points or more.  My account was quite a bit smaller than Rob's as well.

edit:  Portfolio is average of ~9+ months old.
« Last Edit: July 02, 2015, 02:23:51 PM by thezfunk »

lascott

  • Hero Member
  • *****
  • Posts: 1423
    • View Profile
    • Appreciate my post and want to try LendingRobot? URL below
Re: P2P-Picks is off line
« Reply #6 on: July 02, 2015, 02:22:00 PM »
I don't feel like posting all the screen shots unless there is a huge uproar to do so but I can tell you P2P-Picks had me earning ~17% with primary notes only and then I started selling distressed notes on the secondary market so my combined return was ~12%.  My trading only portion was ~4.5% but that was coming up as I started getting out ahead of distressed notes in that instead of waiting for them to go bad I would sell anything with a FICO score drop of 50 points or more.  My account was quite a bit smaller than Rob's as well.
I've only been investing just over a year and have added a lot in the past several months so my account isn't really that telling at this point because my notes are so young. ie. LendingClub Understanding Your Return page: "Weighted Average Age of Portfolio:   5.6 mos". I've also gotten quite a bit more conservative since megacorp did "corporate restructuring" some months back.

If you are an InterestRadar user I would be very interested in seeing your equivalent screen shots.  Nice to get a glimps at the loss rate.
 
Tools I use: (main) BlueVestment: https://www.bluevestment.com/app/pricing + https://www.interestradar.com/ , (others) Lending Robot referral link: https://www.lendingrobot.com/ref/scott473/  & Peercube referral code: DFVA9Y

lascott

  • Hero Member
  • *****
  • Posts: 1423
    • View Profile
    • Appreciate my post and want to try LendingRobot? URL below
Re: P2P-Picks is off line
« Reply #7 on: July 03, 2015, 10:18:58 AM »
I was doing a google search for information related PRs on File.  Look at what was the top google searched. This was before my time so I thought it was a very cool read to see some of Bryce's first comments on his picks!!

Title: Public Records on File.. do you ever invest in the note?
http://www.lendacademy.com/forum/index.php?topic=540.0

I spent the first part of 2012 making the programs and algorithms, and have refined and tested them (with $35k of my own money) this year also. I'm eager to launch a new website service where people will be able to take advantage of my work later in the month--at least a beta version.

It only takes about 2 minutes to make a beautiful Excel sheet with my picks, and then another handful of minutes to make the investment. Couldn't imagine doing it any other way.

When the API comes online I will make a server service that continuously runs my algorithms and automatically invests any idle balances. Or perhaps I will work with others who have developed such systems already. There's some good stuff brewing!

If you get enough loans in which to invest with a strict filter strategy, that's great. It's hard to scale up a filter strategy to sell, though, so that there's enough "product" for all your subscribers.
Many of those questions will be answered on my website within a couple weeks. However, I don't really want to give statistics lessons over the net. A few stats courses and you'd have enough knowledge to do what I'm doing.

By the top 10% thing, I mean that I programmed numerous strategies, including filters, and then compared the returns against each other. I used inception to Q2 2008 to train my modeling and tested all strategies on Q3 2008 to Q2 2009 loans. Their outcomes are known at this time.
Sure, the method will be described in general on the site. It's going to be www.p2p-picks.com, and we're just in the last steps of beta development. Maybe a handful of weeks? It's a side project for me, so it takes a long time to do anything.

The site will have extensive sharing of the results of the strategy, but maybe to the dismay of some people, it will not be couched in a percentage per year / APR standard. I wanted to reduce the number of assumptions in describing the performance of the picks, so I took a buy and hold (and not reinvest payments) approach to calculating everything. It's basically a percentage return over three years. If I had wanted to give a %/year like most people, I would have had to do all sorts of assumptions that have to do with reinvesting payments. As I only wanted to demonstrate that my picks had higher performance than other strategies, I didn't need to do an APR return. I just needed an easily understandable and calculable metric.

I look forward to feedback in a few weeks!
Three years of history and 6000+ completed notes during one of the darkest times in 80 years seemed like a fair training ground to me. People seem to think 6000 is some small number on which you can't base predictions. It is one on which I rest my investment choices, which evolve as more data comes in. Do what you like =). Plenty of room for all comers here.
Let's consider your two propositions:

1) Selection bias in who took out loans in 2008-9. Empirically we have seen principal loss curves improve for more recent vintages (see my guest blog post a few months back), meaning more recent vintages are better. If the best borrowers were selecting into LC in 2009 because they were the only game in town, they sure didn't behave very well. I'll take the more recent vintage performance any day.

2) Things are constantly changing. Some things don't really appear to change very much. I chose variables in my model whose coefficients have remained stable regardless of whatever time period I'm considering. I'd run my models six or seven times, lopping off old quarters each time and many things appeared to have substantially the same relationship to the outcome under study.

Best Wishes!
Tools I use: (main) BlueVestment: https://www.bluevestment.com/app/pricing + https://www.interestradar.com/ , (others) Lending Robot referral link: https://www.lendingrobot.com/ref/scott473/  & Peercube referral code: DFVA9Y

Rob L

  • Hero Member
  • *****
  • Posts: 2060
    • View Profile
Re: P2P-Picks is off line
« Reply #8 on: July 03, 2015, 06:43:45 PM »
I was doing a google search for information related PRs on File.  Look at what was the top google searched.
This was before my time so I thought it was a very cool read to see some of Bryce's first comments on his picks!!
Very cool. Thanks for the history. Before my time too (spring 2013).

So I have to ask myself why I adopted P2PP as my selection method back then:
1) Bryce's resume was/is impressive; particularly with the RAND Corp experience.
2) If I were going to do this I would use this method (logistic regression; industry standard).
3) At the time it's beta, it's free and will do no worse than average if sufficiently diversified. Hopefully much better.
    I've already noted how I screwed this up by adapting $ note size to top%, ignoring diversification impact. Fortunately the impact was small.
4) I've owned my own business for over 25 years and have a policy only to hire people smarter than me. It's served me well.
5) I won't pay someone else 50 bps for something I can do myself. Guess it's an ego thing.
    (ie if I wrote my own model it may be worse than P2PP, but it would not be 50 bps worse than P2PP; Bryce may disagree).
    Make no mistake; 50 bps in this game is a ton. No doubt P2PP provided me at least a 50 bps edge over average as you can see above.