Arbusers

arb12 wrote: ↑

Sun Aug 11, 2024 4:39 pm

Canada's 2024 Global T20,
Montreal Tigers vs. Toronto Nationals Grand Final, hopefully, the game will start as soon as the rain stops.

(1) My last attempts at fine-tuning the 2024 Canadian T20 custom Model weren't too easy due to six eliminated games with "NoResult" or "Abandoned" status.
As a workaround, I decided for today's Model tune-ups to introduce protective margins for quite a few variables. However, due to a lack of time, the range of approximations is mostly based on my observations of the current players and also what I have spotted in the Ontario pitch so far;
(2) Nevertheless, a defined number of variables are handled and calculated in the usual way, based on the common sense of what is seen in contemporary Cricket at similar tiers of the game;
(3) The implied probabilities of the published odds in some of the markets for Tigers vs. Nationals are weird compared to what I've calculated for myself, but I am quite aware of what I am telling you in (1), therefore, choosing an additional grid of well-chosen stop losses would reduce the risk;
(4) I've spotted a lot of things about Andries Gous' behavior on the pitch and decided to move him into a slightly expanded Matrix to try to more accurately calculate Gous's probabilities today;
(5) I think Tim Seifert's sub-optimal spot in the batting order would blow the team against a sudden strong delivery from a fast Bowler (prob >20% today vs the opponent);
(6)A very strong Trigger (prob > 66.7% today) insists that Colin Munro's fairly straightforward updated batting style will produce a clear result in terms of Strike Rate interval of {X-Y, X+Y} against any of the Top Three opposition Bowlers;
(7) Jason Behrendorff's consistency in his supreme Bowling style is among the deciding factors for the outcome, but I wonder what the likelihood of a worsening Economy Rate and an increase in the number of Wckts taken is;
(8) Based on my recent observations, the exact moments of the game that Siddique will turn up to bowl as well as the exact Batsmen he will be competing against are reasonable grounds to try and truncate the appropriate Probability Matrices;
(9) A very quick assessment has to be made of Romario Shepherd's exact form today in his first interactions on the pitch. Why? It has a few unclear personal metrics and that's absurd in its latest sporting state;
(10) The impact of the wet ground;
(11) The ability to use the D/L method when losing overs due to the rain;
(12) And so on.

I'd love to read other Modelers' perspectives. That game is extremely vast in terms of possible states, and that is part of its beauty and unpredictability.

So far, after the first PowerPlay has ended and the Middle Overs have started, some things have gone as expected from the most likely scenarios modeled before the match:
- Tim Seifert is Out, removed by Behrendorff;
- Junaid Siddique came on when the Tigers were in a knockdown-like state (4 Wckts lost within 5 Overs) and he took the fifth Wckt at the very beginning;
- The most expected excellent bowling performance for the Nationals;
- And so forth.

So weird odds posted before the game, the Nationals were rated as underdogs by the odds compilers, giving us good positions to buy at levels above 2 and later to sell large percentages of the positions as close as the Nationals' price was trending to fall below 1.2 and more at the eve of the first Death Overs stage today.
My guess is that the odds compilers were mostly misled by the recent Nationals and Tigers matchups. On the other hand, no one of the protective grid positions I mentioned in the previous posting was in effect today. That overcame my expectations.
Not to mention some of the secondary markets that differ from MoneyLine. My assumptions for some initial mispricing offered by odds compilers were confirmed by those five Bowlers taking 7 Wckts before the Dead Overs and six Bowlers taking 9 Wckts in total before the break time. All the market positions are closed.

Despite my Model showing me a probability of a worsening Behrendorff's Economy Rate today, that didn't happen at all.

Additionally, no matter that the Model awaited the excellent Bowling Dept for Nationals today, I failed to "forecast" that absolutely all the members of the Nationals' Bowling Dept took Wckts today in the Grand Final vs the Tigers.

Despite generating very low probabilities for Colin Munro's last performing style vs that kind of opposite Bowling structure as Nationals, I was amazed at his time-to-live on the pitch, which lasted only two bowling deliveries...


Let's see what happens later in the rest of the Nationals' batting portion of the game.

Although the start of the T20 Aussies @ England clash is still to come due to bad weather in Manchester, let's discuss the ODI Wolves Champions vs Stallions Champions in Faisalabad.

What is the reason that match caught my attention? The presence of players such as Masood, Azam, Mumtaz, Rauf, Zaman, Rizwan, Ahmed etc.

Although Fakhar Zaman played cautiously, waiting for a strong attack, he was dismissed much earlier than expected. Rizwan was fine but Mumtaz was smarter today. Iftikhar was perfect but again faced strong bowling deliveries from Mumtaz. The logical movement of the market odds was influenced by all of the above. I was pleased by the strong Stallions' bowling attack - 4 Wckts taken (Khan) & 3 Wckts (Mumtaz).
But I personally didn't expect Wolves Champions' achievement of 231/All Out after 45 Overs.

By the way, I'm wondering if Mehran Mumtaz is somehow related to another Cricket player Urooj Mumtaz (PAK Ladies in the past).

The second part of the game is now underway and I have tweaked some fine-tuning based on an odd decision about Babar Azam to be put #3 in the batting order.

By the way, the market is not hard at all.

Your opinions?

Well, both Tahir and Babar Azam were dismissed within three Bowling deliveries, and plenty of Dot balls afterward and I think some significant market opportunities for a heavy attack arose at a lower risk for your positions than on similar spotted occasions. However, some stop-loss points also need to be moved according to the right calculated probability of fault.

Addendum: The favorites after the first part of the game (according to the odds compilers, followed by agencies and the Exchanges' offers as well) show us exactly how, in a match involving Pakistan or their domestic match, the historically proven raised probability of a batting collapse should be factored into the modeling on your part.
Thousands of graphics have confirmed that probability and the frequency of appearing for decades.

From live-market Favorites, they became 105/All Out in no time!

What a game!

The 2nd Test India vs Bangladesh in Kanpur, day Fourth in progress now,

After the weak 1st day and consecutive 2nd and third rainy days, some markets aired temporary valuable peaks at last!
Minutes ago Jasprit Bumrah conceded two consecutive Fours vs Miraz and the market prices tended to reach the aggregate region of circa 1.11 - 1.14 for The Draw, the ideal moment for a series of limited Lay The Draw positions due to what is expected by Bumrah. Why limited positions? Due to Kelly Criterion's triggers, bearing in mind we're in a day Fourth, Session 2 and Bangladesh were leading at that moment by 224 R with 6 Wckts lost.
I didn't expect two too quick Wckts taken from Bumrah and one by Siraj (3 Wckts within the next 18 deliveries happened actually), I thought we had a likely interval of at least ten Overs to spot that. That's why the profit was emergency locked via consecutive Back The Draw positions in the region of 1.24 - 1.28 and no open positions on The Draw market after Siraj's Wckt at 72.2 Ov.

After the Storm of Indian batting attacks minutes ago (Jaiswal triple Fours, Rohit Sharma double sixes), my live Model triggered new open positions in a different market due to prices for India falling from 8 to 3 within the desired observed interval (X, Y) today, combined with the current R.R. > 15 at the moment of the (X, Y) forecasts applied and the value of some additional variables I'm interested in.
Here I'll probably hold the positions according to Kelly and don't plan close-outs.

By the way, the graphical curve for India to win almost touched the level of 2 now and bounced slightly back after Rohit Sharma's dismissal...


Anyone live modeling that game?

Due to:
(1) India's apparent switching in the heavy T20I batting mode, trying to accommodate to avoid the draw at home vs Bangladesh;
(2) Having in mind also the circa 40 remaining Overs today and hopefully three full Sessions tomorrow;
(3) Qiick reassesses of India's line-up and their batting order;
(4) Plenty of gathered Stats India/Bangladesh, also the preparation of Kanpur pitch,
it is not hard to calculate the real probabilities for The Draw vs India's Win (currently implied by the market close to 60%:40% respectively).

However, it is much harder for the Model to build forecasts until the 2nd Test ends for factors such as:
- The Weather in Kanpur/pitch conditions;
- As much closer predicted interval as possible in terms of stability of the batting Run Rate's behavior for the India team today and tomorrow;
- Plenty of others, of course.

A couple of minutes ago India leveled the Runs scored in 1st Innings and started to accumulate leading from now on.
According to the market, the probabilities (implied by the odds) for Draw:IndiaWin at the moment of leveling the achieved Runs are 42%:58%, (slightly changed to 44:56 due to the break of the Virat Kohl - KL Rahul partnership, and additionally changed too slightly when Jadeja, Ashwin, Rahul were out, removed by the brilliant Bangladeshi Bowling Dept under the management of Shakib al Hasan (Ex-KKR, SRH)), but I do not agree with the exact percentage tight interval (even more closer to 48%:52% after the IND declaration) before the start of the second innings due to:
- The probability of exhausting the Y number of Wckts in BD's hands in the remaining circa 20 Overs in Session 3 today is corrected. My rationale - reassessed some bowling probabilities about the India Bowling Dept under the management of Bumrah in the next Overs in order to avoid a possible BD's NightWatchman-style in the next remaining Overs today;
- Fast re-build the expected probability of exact batting BD's batting behavior till the end of the day, striving for a Draw at the end of the fifth day.

That's why I see some valuable positions here.

Let's see what happens!

Your opinions?

Apparently, both main graphics curves, describing the initial of the second Innings had included some primary-noise-caused moves till the moment of Ashwin's bowling delivery toward Mahmud in the 9.4 Over.
The forecast of the breaking of the current support/resistance levels would be risky if the secondary-noise originator is not recognized precisely, but that is not too easy before the initial moves of Day 5th.

The IPL 2025,

Teams like Royal Challengers Bengaluru, Punjab Kings (PBKS/KXIP), SunRisers Hyderabad etc. are just excellent in terms of structuring your own observation of the ongoing games and building forecasts, choosing appropriate tune-ups of your current Model, based on Math basis and very deep understanding of what and WHY happened and will happen on the Cricket ground.

Cricket is not just a wonderful game, this Jewel can be described with infinite Variables with the appropriate Mathematical notation, potentially generating very large areas of ​​Probabilities, so the nonsense that odds-makers publish before the match is obvious from afar.

For a very solid reason, I think one of the most appropriate Methods for accurately predicting RCB/PBKS and similar teams in IPL is regression analysis, but today, when I noticed the starting lineups, I decided to experiment with additional intermediate Factor Analysis to help my Prediction Interval (on-the-fly Interval plus final Interval), and those excellent Triggers were golden for the Observation Matrix plus the Error Term Matrix - those were spotted in the Royal Challengers Bengaluru's batting portion of current game vs Delhi Capitals, such as:

..., {Mitchell Starc vs Phil Salt & Virat Kohli: 6,4,4,5nb,6,1,4}, ..., {Vipraj Nigam vs Devdutt Padikkal & Virat Kohli: 1, Double Dot, 1, Double Dot}, ..., {Kuldeep Yadav vs Jitesh Sharma & Krunal Pandya & Rajat Patidar: Dot, Wckt, Dot, Tripple One}, ...
Note, in those moments 4 and 5-digit amounts were vacuumed with ease.

What an incredible beauty of today's IPL game on the Bengaluru pitch, what an unexpected solution to a Mathematical problem on the move from the perspective of ease!

Let's see what happens from now on, Delhi Capitals' Batsmen KL Rahul & Axar Patel vs. RCB Bowling Dept and not only vs Kumar, Sharma & Hazlewood specifically, will they move significantly the current probability interval on the fly or not after the PowerPlay?

I'd love to read different points of view from any IPL Modelers or Cricket observers, thanks!

IPL at Chepauk right now,

Excellent bowling performance from Kolkata Knight Riders' Bowling Department so far,

To calculate some probabilities of the current KKR bowling department in detail, I have been re-analyzing thousands of gathered Teams' and Players' data and Statistics and re-watching videos over the past few days, for example:
- I have re-watched and re-analyzed in detail full videos of past IPL matches in cases where Moeen Ali (currently in KKR) was bowling against the great Chris Gayle at times when the so-called "Gayle-Batting-Storm" (a ton or at least a half-century) happens and almost no one was able to stop him from doing so, and there are subtle details in Moeen Ali's bowling approach when Gayle seriously slows down for a while; The same analyses again, but for other IPL historical games, when Ali was bowling vs Batsmen X, Y, Z and new iterations etc;
- The pre-match calculated likelihood of Moeen Ali doing again these details today vs Chennai Super Kings' opening Batsmen was greater than XX,X % due to the pre-estimated Y factor;
- Calculated Y factor in as tight an interval as possible - note that MS Dhoni is seeded in an emergency as Captain today in a semi-derailed team (by the CSK management, in order to hide behind him) - Ajinkya Rahane and Maheesh Theekshana were released from CSK before IPL 2025, wrong starting lineup from Fleming in the past games -for instance, no Devon Conway vs Royal Challengers Bengaluru, floating Deepak Hooda vs RCB and so on;
- Overusing of CSK Spinners in inappropriate moments, while the game is changing globally and so forth;
- Lack of responsibility for moving MS Dhoni to a higher position in the Batting Order for Chennai Super Kings; I personally would experiment, sending MS Dhoni even at # 3 or # 4 in the Batting Order or so (depending on the current context on the pitch, provided that Dhoni is full of energy at that age);
- Opening pre-match positions, based on hundreds of sheets, full of pre-match analyses and so on;

Lastly, I think KKR and CSK are among the teams (also RR and KXIP) that would hypothetically combine Tymal Mills & Jofra Archer in their Bowling Squad one day in the future. The preliminary modeling of this sounds good so far in my initial set of testings.

At the same time, the PSL 2025 has just started,

Islamabad United vs Lahore Qalandars at Rawalpindi,

I tried to forecast the changing approach of Mike Hesson, but also have Stats & impressions for Andries Gous, Colin Munro, Jason Holder (golden player for that team specifically, guess why?), Shahzad, Wasim etc, but I didn't understand the idea behind the lack of van der Dussen in the starting lineup today???

On the other hand, I'm wondering exactly how will start Fakhar Zaman in the next Qalandar game after today's dismissal and whether the leading partnership in the long term would be exactly Zaman plus Batsmen X or Y for Qalandars in 2025? Why Meredith removed easily Zaman today?
Also, the lack of Kussal Perera & Tom Curran from the starting lineup is changing some probabilities, who knows?

And finally, I think both the PCB and BCCI should rethink the current overlap of IPL and PSL games and reschedule the PSL & IPL calendar to not overlap in the next days, whether the following hypothesis is correct: both federations serve Cricket fans, providing them with the best experience.

From that posting:

post103849/#p103849

and related specifically to this:

arb12 wrote: ↑

Sat Mar 29, 2025 3:41 am

...
For example, Daniel Vettori's coaching preference for a unique heavy batting style at Royal Challengers Bangalore and currently at SunRisers Hyderabad comes with a uniquely volatile Risk/Reward Ratio on the pitch...
Here I mean, do you understand the likelihood of heavy-hitting guns emerging for teams under Coach Daniel Vettori, and under what conditions exactly is arising the likelihood of "unexpected" defeats?
...
SunRisers Hyderabad's attacking heavy guns and unstable Depts are very similar in my opinion to the Dream Team's guns and defense under coach Johan Cruyff, if you remember Barçelona under Cruyff, it produced endless streaks of results like 8:2, 7:2, 6:3, 5:4, 4:3, 6:2 and so on. But also drawdowns 0:4, 2:4, 3:6 - full enjoyment for the watchers, and also due to the implementation of Daniel Vettori's vision, in the modern IPL we today often watch Masterpieces such as 286/6 : 242/6 vs Rajasthan Royals, that notable SRH's 287/3 : 262/7 vs Royal Challengers Bengaluru in IPL 2024 (those ODI-like results are in T20 format, therefore under heavier risk) and similars (Vettori's teams historically turned the achievement of over 200 runs a norm), but when the drawdowns happen, such as "unexpectedly weak" final vs Kolkata Knight Riders in the IPL 2024 Big Final or yesterday's starting as a big favorite vs the LuckNow SuperGiants and "unexpected" defeat 190/9 : 193/5 vs the Rahul Dravid's team in the end?
...
Results in the short-term/medium-term at mitigated risks? Try to model the KXIP/PunjabKings, and their real pre-match probabilities vs the offered ones will amaze you. Their things have been the same for years with short-term variations and unique supports/resistance levels and when the pendulum in those unique top KXIP team from the past (Gayle, Ashwin, KL Rahul, Curran, Pooran, Miller, Shami etc, under Mike Hesson) was at position XX1, years later the pendulum will highly likely (two Sigmas probability) move in the range of the XX1±Y, but inside that interval, when very specific pre-defined conditions arise in-live mode, they often overcome pre-match favorites from unreal positions as well as they failure vs underdogs in a mystical way only spotted in David Lynch's scariest films. The agencies often offer unreal odds before Punjab Kings games. Let them prolong those kinds of offers.
...
When it comes to modeling the up-to-date 2025 KXIP/PunjabKings, the Circle of Horror is closing with Ricky Ponting at the helm!
...

Kings XI Punjab @ SunRisers Hyderabad at the moment,

Naturally, Ricky Ponting vs Daniel Vettori is a pretty clear game every time they meet, as passionate Cricket on the pitch as a shiny sunrise near Barbados;
Batsmen's cannons are running non-stop, as usual;

When PBKS decided to bat first today, it was pretty obvious that they would attack the current top ten results so far in the IPL Power Plays. I remembered the Ponting vs Vettori classic from IPL 2024 when the PowerPlay was 125/0 in favor of the SunRisers vs the DareDevils!!! (also the collapse of Delhi that day from 199/6 to 199/All Out, OMG!!!)

The problem here was this: no agency offered exactly what I wanted to put on the KXIP/PBKS@SRH markets (to attack the historic Top Ten IPL PowerPlay Rank with my own calculated odds), so I had to find a workaround through weird cross-markets across multiple oddsmakers to make a complex market position that was something like what I wanted; All that combinations were possible due to the fact, that Cricket markets are like Forex/Stocks - endless liquidity and full of derivatives.

Shreyas Iyer's Six @ 5.6 Ov secured 89/1 for PBKS in 1st PP today and displaced 88/1 by Kolkata Knight Riders vs Delhi Capitals from IPL 2024 (Ricky Ponting was involved here, again);
By the way, today PBKS were 53/0 after 3rd Ov (six Fours & four Maximums), while Knight Riders were 32/0 at that point (five Fours & zero Maximums), before acceleration in the 2ndHalf of the 1stPowerPlay - recall those famous {Double Max, Four, Dot, Max, Four} by Sunil Narine in the next Ov, ended in total {Narine+ Raghuvanshi: eight Fours & four Maximums} after the 1st PP and that was achieved outside the Eden Gardens!!!

I expect a lot of interesting things to come up in the chase over the next two hours in Hyderabad, would you please write your views and opinions?

By the way, at the same moment, Mohammad Rizwan achieved a Half-Century within 38 deliveries for Multan Sultans & Karachi Kings, please, comment also on the PSL.

This overlap of IPL and PSL causes many more screens to be included in front of the Observers.

We will finally see a Super Over in the IPL 2025,
In my view, if Sanju Samson didn't retire hurt in the batting portion for Rajasthan Royals earlier today, the Super Over wouldn't be possible.
Your points of view regarding Rajasthan Royals @ Delhi DareDevils?

Slightly off topic but still statistically related to the market entries today,
Watching with great interest Eshan Malinga's bowling performance for SunRisers Hyderabad today vs Mumbai Indians at Wankhede, the home of his uncle Lasith Malinga. Memories and recollections of the monstrous bowling performance of the previous SunRisers Hyderabad team (during the Kane Williamson era!!!) and those memorable Mumbai Indians' Wckts taken by Lasith as well as the impossible to any other human being unique Economy Rate achieved of the bowling ​​deliveries by Lasith Malinga in those famous IPL finals during the golden era...
Your opinions?

IPL, Kings XI Punjab @ Royal Challengers Bengaluru just started about 160 minutes after the original toss time due to rainy conditions,
Re-analysing the pitch conditions and starting lineups has led me to open some additional calculated cross-market positions to shape what is most likely to happen on the pitch under these conditions, but if the umpires shorten the overs due to wet conditions later in the game, the additional new positions will be computed.
Meanwhile, PSL is in full swing, Quetta Gladiators @ Karachi Kings did not offer big reasonable offers at the desired pre-calculated odds to be placed relatively risk-free before that match, but in my opinion, sometime towards the end of the 1stHalf of the 1st PowerPlay the primary market, as well as the secondary markets, gave clear technical indicators of the upcoming mismatch between a very risky attempt
of a sudden take-off near the end of the 1stPP by the opening duo Batsmen of KK:{David Warner & Tim Seifert} against the lurking bowling predators of QG:{Abrar Ahmed & Saud Shakeel}, so when the predicted mismatch happened in {Double consecutive (Double Fours, One) by Seifert} which was the reason Warner was trapped by Ahmed plus the over-excited and nervous Warner being trapped easily by Shakeel, guess why, so the sum of two Wckts for 5 deliveries was probably the first ground in order to place new reasonable entry/exit points of the live markets today, The next triggered market positions were indeed caused by the renewed fundamental analysis, i.e. when Sean Abbott returned to the bowling attack to bowl his second Over around the Middle Overs stage.
And so on.