We've been receiving a tremendous number of e-mails (not to mention forum posts) with comments/ideas/criticisms on XM Exchange. We're thankful for all feedback we receive and I apologize that we haven't had the opportunity to respond to most of the correspondence we've received. But I do want to emphasize that we're definitely taking notes.
I'm creating this thread so in the hope that we can consolidate player feedback on the Exchange in a single location. Hopefully this will become kind of a player wish list as well.
Following are some of the most frequent and interesting suggestions and criticisms we've received (there's been a lot of overlap so I've compiled many related themes into single items). I've also included in many of the cases a bit about out plans to address these issues.
I'm hoping to see this list continue to grow even as we start knocking items off one by one.
So if any ideas please please please post them here.
Comments and Requests We've Received:
- Players have complained about point spread/totals markets that deviated substantially from widely available numbers. A lot of players just don’t want to be betting in markets with stake numbers 2 or 3 points off the Vegas figure. We’ve addressed this issue for the time being by having improved the speed and reliability of our automation process for pulling new spreads and totals from line services, which we then use to automatically create new markets.
- The downside of this, brought up by a number of players as well, is that many events eventually end up with a completely unwieldy number of market figures, both drying up liquidity and over complicating the interface.
We do plan to address both these opposing concerns in the near future by offering players the ability to place “linked” offers across multiple spread or total markets, all of which draw on a single capital source. XM Exchange players will have the ability to place a sell offer in a single market and then optionally specify additional half point buy/sell prices which will cause their single offer to appear in any number of additional markets (at the player’s option, of course).
In this manner a seller would only have to manage a single offer at a single line, but still have the benefit of a presence in multiple markets. He’ll increase his chances of a fill, providing liquidity to multiple “nearby” lines without tying up any more capital than that required by a single order.
So for example a player could simultaneously offer $100 in risk on each of UCONN +1, UCONN PK, and UCONN -1, all entered (and modified as desired) as a single order at either user specified or default point buy/sell prices, with each one decrementing in size as the other orders are filled. So you’d have $300 of offers outstanding, but would only be tying up $100 in bankroll.
If you can buy and sell half points at sportsbooks why shouldn’t you be able to the same at exchanges?- Offerings in parlays and teasers. Similar to the above, market makers (and indeed ANY interested player) will have access to a simple interface that will facilitate efficient liquidity provision in compound markets. This would be based on just a small number of input variables (with default values available) and would obviate traders of the burden of monitoring anything other than the associated underlying markets. For players just looking to bet parlay this should be completely transparent and fundamentally no different than what he’d find at a traditional sportsbook.
- Grading time – initially the time to settle wager was unacceptably slow. In truth this was because we were then part way between manually entering scores and automating score entry via what turned out to be a rather poor data source. After a couple of missteps with similarly slow and unreliable feeds we’ve put in a place a fast and accurate data solution that submits scores as soon as they’re received to the grading engine, which then settles all such wagers immediately. This is just fundamental
- When buying across multiple prices, players have expressed a desire for the software to automatically tally up all the risk quantities at intervening prices. This way if there were say, $50 for sale at +110, $20 for sale at +109, and $44 for sale at +105, the user wouldn’t have to mentally sum these dollar values, rather the software would automatically enter the buy of $114 at +105 thereby clearing out all those prices with a single mouse gesture.
- Further customization of site settings (persistent across user sessions) including: skinning (customizable color palettes); optional inclusion of betting tool widgets that may be repositioned anywhere within a given player view; customizable game and event ordering that may be based on such criteria as preset user “favorites”, degree of market activity, proximity to game time, the player’s volume of live and/or filled order in that market, as well any of a number of additional sets of event grouping criteria that players might suggest. I happen to think the event panels are great, but they’ll be even better when they can be further persistently customized per user.
- Related to the above would an improved navigational system that the obviated the need to traverse multiple widget levels to enumerate/locate events of interest. This night be accomplished through a cascading menu interface similar to BetFair, a Microsoft Office-type “ribbon bar into which particular markets can be dragged, extruders located at the four sides and four corners of the each providing easy access to a particular sport or league as defined by the user.
- Ability to view not only individual fills per market, but also fills grouped by price, by side, as well as net position across an entire market. A proof of concept-type version of the latter can be found at http://www.xmexchange.com/static/html/net_fills.html. (Player must be logged in to use it, the software is unofficial, unsupported, and only mentioned here because so manhy players have asked for it). We’ll have a much a fuller featured version directly integrated into the UI soon.
- Ability to sort fills and open orders not only by market side, but also by event, risk/win amount, net risk/win across markets, fill time, and whatever else.
- Option to view lines in any of the major formats.
- The ability to queue multiple orders for eventual simultaneous submission. Enter the orders now but then submit them all at once later, Matchbook has of course had this for a while … but our players will be able to do this over any number of disparate markets and events.
- Single-click order cancellation by price, side, market, event, or across all a player’s offers. Single click order submission (obviously OFF by default, and limited mindful of the user defined single click size limitations).
- Timed orders expiry … . Right now, the only automated cancellation occurs shortly before game time. There\s no reason players shouldn’t be able to additionally customize these times as they see fit -- per-order, in bulk, based on absolute time, or based probased on proximityu to game time.
- At the option of the player, automated removal of stale offers (away from the market) and automated parceling of an order throughout the course of the day.
- Instant messaging of binding offer requests – say one player is trying to buy at +102, someone else is offering to sell at -102. Players have expressed a desire for the ability to message the seller at -102 and attempt to secure a middle trade at +100. The order message would be binding on the player who sent it and would only be valid for a brief user-specified period of time (maybe around 30-60 seconds by default). Players could choose to identify themselves when making and receiving such offers or could remain anonymous. And obviously players would need the option to switch off the receipt of such messages
- User created markets and events – Matchbook had this with their X-Bets offerings but it was slow and tedious and required e-mailing their CS. We’re going to have a simple interface such that games or events already listed on major line services or perhaps at certain sportsbook will be added by request almost immediately. Custom events and props will still need to be approved by staff for suitability and clarity, but it will be handled through a site-integrated interface. I already mentioned how we\re going to be handling additional many spreads and totals automatically, but even in situations when a particular market doesn’t happen to fall within anyone’s range players will be add whatever as they desire (although with some reasonable sanity checks).
- More flexible usage of trading P&L across linked markets, and the removal of usage restrictions on trading P&L coming from events that are always ACTION.
- Improved tutorial in terms of clarity, navigability, breadth, and organization,. Related to this is as a “guided usage mode” for new player.
- The ability to filter from view orders less than that of a specified size.
- And finally, the all-important API:
I certainly know as well as anyone that a betting exchange requires a robust, well-documented, and well supported API. While of limited to interest to most players, this was still probably the single most inquired about feature. On a personal note I spent most of my professional life as a quantitative trader, over the years developing, implementing, and trading all flavor of automated trading strategies on a variety of proprietary platforms.
As such I can’t understate the level of importance I attribute to the continuous development of, and informed top tier support for, our API product. To take this a step further, our intention is to develop and maintain open-source API “helpers tools” (in-house) structured to appeal to players that cut a wide swath across degrees of technological sophistication. This should be of great benefit to highly profit-oriented players who might nevertheless lack in sufficient technological resources to develop their own automated trading platforms.
What we figure is that the easier we can make it for a player to trade, the more relevant, useful, and intuitive tools we can offer him -- the more attractive the trading environment will be for liquidity providers.
So not only will we be supporting the API itself, but also assisting players like never before in bridging the gap between their Excel spreadsheets/proprietary databases, etc. and direct and seamless order submission to the exchange via our API.