This seems like a very weird pricing strategy on the one hand and a very weird charitable strategy on the other.
Frankly, the last thing poor people need is free access to high-end stock trading tools so they can piss away their money on losers and transaction fees.
This seems about the same as offering poor people free rides to the gas station so they can buy a lottery ticket. I mean, you're giving them a free ride, but to what end?
I'm the founder of Tiingo and just saw a huge influx of users! Thank you HN!
I 100% agree with you. The podcasts are there for educational purposes and I try my best to emphasize the limits of quantitative modeling and backtesting. I was a former quant trader and saw experienced people make the same mistakes.
My goal on the site is therefore give tools, but not recommendations. It's the reason I don't associate with brokerages either, I want the data to be unbiased as possible.
My hope is that newcomers and experienced individuals alike come onboard and listen to the podcasts that emphasize just this.
In terms of pricing strategy, I will be enforcing a minimum $1/month floor coming in the next couple weeks.
But either way, people deserve so much better and I'm going to try my best to make sure the tools and information remain intellectually honest.
-Rishi
I'm a current student so please forgive my naiveté of any real-world implementations.
You mentioned "My goal on the site is therefore give tools, but not recommendations."
Do you have any plans to offer portfolio construction tools? I took a portfolio construction class which used Andrew Ang's book as a theoretical basis and was struck at how simple it appeared to create [insert weighting scheme here] portfolios.
I think though a platform could go a step further and demonstrate how to execute different strategies that appear to be only executable by hedge funds. i.e. can you show [insert publicly-tradable assets] which when bought together appear to replicate a hedge fund's beta (maybe not HFRI but some sort of market-neutral beta)?
I dunno it just feels there's a lot that can be done to remove the mysticism from wealth management and open data is just the beginning of such a movement.
E-trade arguably does just as much damage, letting an average end user (who often lacks the industry experience of both the financial industry as well as the company in which they choose to invest) throw their money into anything. A VC once said he knew the Web 1.0 boom was over when his cabbie told him to invest in Cisco.
I, for one, think that the Bloomberg reign -of-terror can't end soon enough. They're as bad as Elsevier in monopolizing information. Information can be used properly or improperly in any context[1]. You want to see some real damage? Give that demographic access to Q/KDB+ and let them underwrite a 20x inter-day margin with their mortgages.
[1]Go to any o-chem forum and you'll see tons of graduate students talking about research chemical drug synthesis techniques, precursor availability, ways around DEA watched-chemical lists, etc. It's vaguely masked in their own lexicon, but it's plain as day.
Edit: By throwing away ones 401k, I meant "cashing out their 401k and effectively gambling on a handful of stocks". Low-load / no-load 401k's and (Roth) IRAs are way safer prospective investments often with useful tax benefits. I keep my 401k with Vanguard and they explicitly say "Very few Vanguard funds charge fees when you buy and sell shares. The fees are designed to help those funds cover higher transaction costs and protect long-term investors by discouraging short-term, speculative trading." which is a mentality that most people should adopt. https://investor.vanguard.com/mutual-funds/fees
I don't understand this opinion, and want to.
Bloomberg work with 3rd party data that's licensed not only to them, but multiple other platforms too. Elsevier monopolise sources so they're the only publisher.
A lot of data on Bloomberg is public, just organised in a really familiar (to Bloomberg users) UI.
Getting into detail: A lot of data in markets is simply inaccessible on any platform (hidden orders, etc), and some data services try to discover this, but that's not like an academic journal monopolising papers.
Not really? I mean, sure, maybe a few are, but on the whole 401(k) returns are actually shockingly good considering the level of control people have over them. Default target-date funds work really well.
The issue I have with pay-what-you-want or low revenue business models is they're unsustainable.
Either declare something a passion/charity project and make it complete free (but perhaps open to receiving donations) to properly set users' expectations or monetize it.
I takes a lot of time to develop and maintain complex technology like trading tools.
I don't know what the founder's background is. However, if he makes good money from other sources, after the initial euphoria of sharing his tools with the world goes away, he'll have stay motivated doing the grunt work to maintain it. And pay out of pocket for expenses and the value of his own time a PWYC pricing model likely won't cover.
Making money (and a profit) isn't evil or bad, it's necessary both to see if your market (and audience) deems your product worth developing and to keep yourself motivated (and compensated fairly enough) to work on something.
A lot of companies have shut down due to lack of funds (or the ability of makers to go without making much money). It's risky for an user to try a product with a tenuous future, come to base their work flow on it only for it to be closed down down the road. Especially with something like trading.
To address this concern outright: in the next couple weeks I was (and still am) planning to make a minimum payment of a $1/month. Without divulging too much into why, the company can be wildly profitable at such a price and my breakeven is very low.
My background has been specializing in structuring data, scalable computing, and also building trading software and trading models/algos.
I'm in the middle of a raise and so far all the investors, when seeing how I was able to pull this all off, are onboard with the pricing model and are eager to get started working together.
I can't divulge too much further publicly quite yet- but not only is this pricing model feasible, but has gotten the approval of the small group I've demonstrated it to.
Like I said, expect a $1/month minimum going forward in the next couple weeks. Also, I've been doing this almost 2 years now, full-time and without a paycheck, and the euphoria hasn't wore off :)
Having investors changes the equation quite a bit. Depending on how much they've put in, that should give you a good runway to be able to experiment with the pricing model. Wish you the best of luck!
I'm a huge data integrity nut, a spillover from when I used to trade quantitatively. I run my own data cleaning algos and I've found Sharadar's data to be far cleaner than sources like bloomberg.
A lot of people assume BBG's (bloomberg) data is impeccable, but I had an entire database of mistakes I found in Bloomberg that I had to correct. Any quant trader will tell you this. In the middle of the day, Bloomberg would swap the 2nd and 3rd continuous futures contract. It drove me nuts when it happened and was a motivation behind Tiingo.
There is this false sense that financial data needs to be expensive. No - to me the public data is a commodity and that's the way the world of financial data is moving. Vince shares this idea with me and encourages me to be more open with my data. I will be offering an API pilot program in the coming week as I develop my own API for the data I source myself.
Vince's company is: http://sharadar.com/
And is available via the Quandl API http://www.quandl.com
Also, not only does Tiingo source its own dividend data, but it shows its work. if you go to https://www.tiingo.com/d/t and hover over the binoculars you will see the values highlighted.
I do this to fight the idea of perceived value when it comes to financial data. Nobody else will give you this level of detail for dividend data. I started doing this because I found my existing dividend vendor data riddled with errors. That's how nutty I am about data.
I regularly (at least once a week, sometimes multiple times per day) BBM'd the Bloomberg helpdesk with notifications that their data was wrong. This varied across many of our needs, but even simple stuff like money supply / macro stats was often just wrong. How had no one spotted this before? We ended up just using Datastream terminal for macro data.
Killer feature is Excel plugins. Do you have plans for this?
Edit: Just tried Quandl didn't know about this before. What a fantastic tool.
Very impressive bio btw.
Having a link to those numbers as supporting backup would be convenient, such as the 10-K.
I think it is nice to have a quick confirmation, as well as reading other information that the quote provides (footnotes, prospectus, 8-K)
