Ritchie - Web Site Update

[kmiklas]

You're site says exchange but your concept looks a lot more like some sort of ATS internalizer from what little info there is.

The IEX is the 13th exchange in the U.S.. It took an enormous amount of regulatory red tape and VC from nearly a dozen large companies. That is a legit lit exchange, what I think of when I saw your claim of being an exchange.

You talk of being fair, however I see the IEX model actually making a level playing field. For example the IEX places price (bids & offers) over queues which have been gamed to death by spoofing/layering on the other exchanges. They have a built in delay to prevent predatory front running HFTs from operating whereas the good market making HFTs like this. They do not allow collocation, in fact you have to be at least 42 miles from their exchange.

Maybe I just didn't understand what you are doing. Perhaps this is more about providing tick data for the API users and not about matching buyers with sellers.

Thank you for this comment. It is very helpful.

I have not looked at the IEX (I will now!) but this business of introducing delay concerns me. I'm really trying hard on Ritchie not to introduce lag or delay: I want it to be supersonically fast. The key is to make it supersonically fast for everyone, not just for those who can afford a 1ft direct fiber connection from an ASIC chip to the exchange!

That being said, I will read up on the IEX model and get back to you.

 

[comagnum]

The IEX delay is negligible to order fills at 350m/us - just a tad slower then the speed of light (299.792m/µs) but hoses up the bad HFT's from being able to use a latency edge to front run orders. The NYSE has also adopted this. A lot of the big well established futures traders like the national cattleman have been asking for the CME to put in a delay as well - in fact they have threatened to stop trading without it.



http://pointsandfigures.com/2016/01/23/the-national-cattlemen-say-markets-are-broken/

 

[MrMuppet]

Wow, wtf did I just read? You guys are planning your own exchange??

However, I'm not quite sure if you have the right background/knowledge or if you work on it from the right angle...

On the one hand, making an exchange "supersonically fast so everyone has the same speed" is nonsense. The speed race is not about co location anymore, the arb - and queue game is about min/maxing distance to all exchanges, i.e. you want to lift offer/cancel offer in 500 stocks when the ES upticks, you need optimal speed between NYC and Chicago.

Making it a superfast and "co location for everyone" exchange is just summoning the pay to play crowd with their 100M$ infrastructure the small guy cannot compete against.


On the other hand, the money is made with rebates, HFT's get paid for providing liquidity. The free/inverse exchanges are super cheap. Richie is asking for 25cts per 100 (!!!).

Even as a retailer you can get 10cts per 100 and I don't even talk about prop shops here.

All the sub - penny trading is about 0+ strategies, meaning that you buy at 10.01001 leaning on an order at .01, sell it at .02 and collect the rebate for the limit order. The risk reward for that is 1:100+ depending on the rebate. If you ask for 25cts/100, you bring that down to 1:75.

In addition, what about ordertypes? FOK, ISO, AON, Hide/not slide, Midpoint, Midpoint Peg? Or just LMT? There are almost 200 order types in the US - Stock market.

What's the point of Richie? Will you position yourself as a retail - ATS?

Before you keep spending further time on the project, I'd suggest you stop right here, download all exchange rulebooks and educate yourself. It seems like you know nothing about market microstructure or the actual problems of fragmentation. The IEX case was the biggest thing since ISLD and you don't know about it??? And you are building an exchange???????

I really don't want to bust your bubble but I figure that you would do way better, if you gathered some knowledge to find a niche for Richie that actually makes sense.

Even if you manage all the hardship and get an eagle from the SEC, your biggest problem is yet to be solved: How do you incentivise people to trade on YOUR exchange instead of all the others? How do you attrack volume? At this point in time, all I can see is a fancy matching algo and cheap data/co - lo...unfortunately this wont cut the cheese

 

[MrScalper]

Re: Data feed to Excel, I see now. You just want a simple Dynamic Data Exchange (DDE) interface to the feed. I thought that you wanted to gather historical data.

I plan to offer you the ability to buy and sell a financial instrument for one U.S. quarter (US$0.25) per trade. Free tick data, and total visibility to book data... none of which is delayed or a VWAP.

Will keep an eye on your progress..$O.25 seems OK for certain markets..if you get Excel link working let us know as I will definitely look at it.. Cheers

 

[MrScalper]

Wow, wtf did I just read? You guys are planning your own exchange??

However, I'm not quite sure if you have the right background/knowledge or if you work on it from the right angle...

On the one hand, making an exchange "supersonically fast so everyone has the same speed" is nonsense. The speed race is not about co location anymore, the arb - and queue game is about min/maxing distance to all exchanges, i.e. you want to lift offer/cancel offer in 500 stocks when the ES upticks, you need optimal speed between NYC and Chicago.

Making it a superfast and "co location for everyone" exchange is just summoning the pay to play crowd with their 100M$ infrastructure the small guy cannot compete against.


On the other hand, the money is made with rebates, HFT's get paid for providing liquidity. The free/inverse exchanges are super cheap. Richie is asking for 25cts per 100 (!!!).

Even as a retailer you can get 10cts per 100 and I don't even talk about prop shops here.

All the sub - penny trading is about 0+ strategies, meaning that you buy at 10.01001 leaning on an order at .01, sell it at .02 and collect the rebate for the limit order. The risk reward for that is 1:100+ depending on the rebate. If you ask for 25cts/100, you bring that down to 1:75.

In addition, what about ordertypes? FOK, ISO, AON, Hide/not slide, Midpoint, Midpoint Peg? Or just LMT? There are almost 200 order types in the US - Stock market.

What's the point of Richie? Will you position yourself as a retail - ATS?

Before you keep spending further time on the project, I'd suggest you stop right here, download all exchange rulebooks and educate yourself. It seems like you know nothing about market microstructure or the actual problems of fragmentation. The IEX case was the biggest thing since ISLD and you don't know about it??? And you are building an exchange???????

I really don't want to bust your bubble but I figure that you would do way better, if you gathered some knowledge to find a niche for Richie that actually makes sense.

Even if you manage all the hardship and get an eagle from the SEC, your biggest problem is yet to be solved: How do you incentivise people to trade on YOUR exchange instead of all the others? How do you attrack volume? At this point in time, all I can see is a fancy matching algo and cheap data/co - lo...unfortunately this wont cut the cheese

Your arguments make sense..and you might be right..but you might also be wrong!

The main point is that he might actually come up with a product that certain people will use..for their own reasons..as to the size of this market..and it possible revenues..that of course is the big question.. nothing ventured nothing gained..I have seen a lot worse things that are still up and running..so they must be making some money!

 

[MrMuppet]

As said, it was not my intention to bash him, but I think constructive criticism should not always be unicorns and roses.

He might be up to something but it was my impression that he picked up some lines from the newspapers about HFT and thought "Hey, lets make the world a better place". Fair enough.

But The US - Equity market is such a complex monster that IMO you should know what you are doing and why. As a trader and especially when you want to create something ambitious like an exchange.

You are up against the very (very!!!) big guys, companies that have massive stakes in lit exchanges as well as their own ATS's. And whenever somebody shows up to chip away at their volumes, they become very nasty very quickly.
Or what do you think will UBS do when some startup cuts into their darkpool volume? Will they sit there and applaude or rather unleash all their legal departments power? Read the story of IEX and how much hassle it was for those guys to get approved by the regulators.

In terms of costs, remember that 25cts/100 is just the exchange fee. Brokerage fees, NFA fees and Clearing Fees could expand your half turn to way above 1$/100...and honestly, nobody but the mom and pop gamblers will pay that.

But Aunt Sally won't make markets -> no liquidity -> no trading - > no volume -> no money for Richie...simple as that.

https://iextrading.com/trading/alerts/2016/036/
Look at the fee schedule here...besides executing undisplayed orders and routing, everything is free.

I'll have an eye on it, but unless there is a massive improvement in terms of pricing AND structure/rulebook, I don't think I will bite. All the guys who are excited for this are small fry algo traders who look forward to cheap data and co - lo, but you need the big firms to trade on your venue.